• Asencio, N., , J. Stein, , M. Chong, , and F. Gheusi, 2003: Analysis and simulation of local and regional conditions for the rainfall over the Lago Maggiore target area during MAP IOP2b. Quart. J. Roy. Meteor. Soc., 129 , 565586.

    • Search Google Scholar
    • Export Citation
  • Bougeault, P., and Coauthors, 2001: The MAP Special Observing Period. Bull. Amer. Meteor. Soc., 82 , 433462.

  • Bousquet, O., , and B. F. Smull, 2003a: Airflow and precipitation fields within deep Alpine valleys observed by airborne Doppler radar. J. Appl. Meteor., 42 , 14971513.

    • Search Google Scholar
    • Export Citation
  • Bousquet, O., , and B. F. Smull, 2003b: Observations and impacts of upstream blocking during a widespread orographic precipitation event. Quart. J. Roy. Meteor. Soc., 129 , 391409.

    • Search Google Scholar
    • Export Citation
  • Braun, S. A., 2006: High-resolution simulation of Hurricane Bonnie (1998). Part II: Water budget. J. Atmos. Sci., 63 , 4364.

  • Braun, S. A., , R. Rotunno, , and J. B. Klemp, 1999: Effects of coastal orography on landfalling cold fronts. Part II: Effects of surface friction. J. Atmos. Sci., 56 , 33663384.

    • Search Google Scholar
    • Export Citation
  • Browning, K. A., , F. F. Hill, , and C. W. Pardoe, 1974: Structure and mechanism of precipitation and the effect of orography in a wintertime warm sector. Quart. J. Roy. Meteor. Soc., 100 , 309330.

    • Search Google Scholar
    • Export Citation
  • Bruintjes, R. T., , T. L. Clark, , and W. D. Hall, 1994: Interactions between topographic airflow and cloud/precipitation development during the passage of a winter storm in Arizona. J. Atmos. Sci., 51 , 4867.

    • Search Google Scholar
    • Export Citation
  • Colle, B. A., 2004: Sensitivity of orographic precipitation to changing ambient conditions and terrain geometries: An idealized modeling perspective. J. Atmos. Sci., 61 , 588606.

    • Search Google Scholar
    • Export Citation
  • Colle, B. A., 2008: Two-dimensional idealized simulations of the impact of multiple windward ridges on orographic precipitation. J. Atmos. Sci., 65 , 509523.

    • Search Google Scholar
    • Export Citation
  • Colle, B. A., , and Y. Zeng, 2004: Bulk microphysical sensitivities within the MM5 for orographic precipitation. Part II: Impact of barrier width and freezing level. Mon. Wea. Rev., 132 , 28022815.

    • Search Google Scholar
    • Export Citation
  • Colle, B. A., , and S. E. Yuter, 2007: The impact of coastal boundaries and small hills on the precipitation distribution across southern Connecticut and Long Island, New York. Mon. Wea Rev., 135 , 933954.

    • Search Google Scholar
    • Export Citation
  • Colle, B. A., , C. F. Mass, , and K. W. Westrick, 2000: MM5 precipitation verification over the Pacific Northwest during the 1997–99 cool seasons. Wea. Forecasting, 15 , 730744.

    • Search Google Scholar
    • Export Citation
  • Colle, B. A., , J. B. Wolfe, , W. J. Steenburgh, , D. E. Kingsmill, , J. A. Cox, , and J. C. Shafer, 2005: High resolution simulations and microphysical validation of an orographic precipitation event over the Wasatch Mountains during IPEX IOP3. Mon. Wea. Rev., 133 , 29472971.

    • Search Google Scholar
    • Export Citation
  • Cox, J. A. W., , W. J. Steenburgh, , D. E. Kingsmill, , J. C. Shafer, , B. A. Colle, , O. Bousquet, , B. F. Smull, , and H. Cai, 2005: The kinematic structure of a Wasatch Mountain winter storm during IPEX IOP3. Mon. Wea. Rev., 133 , 521542.

    • Search Google Scholar
    • Export Citation
  • Durran, D., , and J. Klemp, 1982: On the effects of moisture on the Brunt–Väisälä frequency. J. Atmos. Sci., 39 , 21522158.

  • Garvert, M. F., , B. A. Colle, , and C. F. Mass, 2005a: 13–14 December 2001 IMPROVE-2 event. Part I: Synoptic and mesoscale evolution and comparison with a mesoscale model simulation. J. Atmos. Sci., 62 , 34743492.

    • Search Google Scholar
    • Export Citation
  • Garvert, M. F., , C. Woods, , B. A. Colle, , C. F. Mass, , and P. Hobbs, 2005b: 13–14 December 2001 IMPROVE-2 event. Part II: Evaluation of cloud and precipitation structures in the MM5 model simulation. J. Atmos. Sci., 62 , 35203534.

    • Search Google Scholar
    • Export Citation
  • Garvert, M. F., , B. F. Smull, , and C. F. Mass, 2007: Multiscale mountain waves influencing a major orographic precipitation event. J. Atmos. Sci., 64 , 711737.

    • Search Google Scholar
    • Export Citation
  • Georgis, J-F., , F. Roux, , M. Chong, , and S. Pradier, 2003: Triple-Doppler radar analysis of the heavy rain event observed in the Lago Maggiore region during MAP IOP2b. Quart. J. Roy. Meteor. Soc., 129 , 495522.

    • Search Google Scholar
    • Export Citation
  • Hill, F. F., , K. A. Browning, , and M. J. Bader, 1981: Radar and raingauge observations of orographic rain over South Wales. Quart. J. Roy. Meteor. Soc., 107 , 643670.

    • Search Google Scholar
    • Export Citation
  • Houze, R., , and S. Medina, 2005: Turbulence as a mechanism for orographic precipitation enhancement. J. Atmos. Sci., 62 , 35993623.

  • Janjić, Z. I., 1994: The step-mountain eta coordinate model: Further developments of the convection, viscous sublayer, and turbulence closure schemes. Mon. Wea. Rev., 122 , 927945.

    • Search Google Scholar
    • Export Citation
  • Kain, J. S., 2004: The Kain–Fritsch convective parameterization: An update. J. Appl. Meteor., 43 , 170181.

  • Kirshbaum, D. J., , and D. R. Durran, 2005: Atmospheric factors governing banded orographic convection. J. Atmos. Sci., 62 , 37583774.

  • Koch, S. E., , B. S. Ferrier, , M. T. Stoelinga, , E. J. Szoke, , S. J. Weiss, , and J. S. Kain, 2005: The use of simulated radar reflectivity fields in the diagnosis of mesoscale phenomena from high-resolution WRF model forecasts. Preprints, 32nd Conf. on Radar Meteorology, Albuquerque, NM, Amer. Meteor. Soc., J4J.7. [Available online at http://ams.confex.com/ams/pdfpapers/97032.pdf.].

  • Marwitz, J. D., 1987a: Deep orographic storms over the Sierra Nevada. Part I: Thermodynamic and kinematic structure. J. Atmos. Sci., 44 , 159173.

    • Search Google Scholar
    • Export Citation
  • Marwitz, J. D., 1987b: Deep orographic storms over the Sierra Nevada. Part II: The precipitation processes. J. Atmos. Sci., 44 , 174185.

    • Search Google Scholar
    • Export Citation
  • Medina, S., , and R. Houze, 2003: Air motions and precipitation growth in alpine storms. Quart. J. Roy. Meteor. Soc., 129 , 345371.

  • Medina, S., , B. Smull, , R. Houze, , and M. Steiner, 2005: Cross-barrier flow during orographic precipitation events: Results from MAP and IMPROVE. J. Atmos. Sci., 62 , 35803598.

    • Search Google Scholar
    • Export Citation
  • Medina, S., , E. Sukovich, , and R. A. Houze Jr., 2007: Vertical structures of precipitation in cyclones crossing the Oregon Cascades. Mon. Wea. Rev., 135 , 35653586.

    • Search Google Scholar
    • Export Citation
  • Rotunno, R., , and R. Ferretti, 2001: Mechanisms of intense alpine rainfall. J. Atmos. Sci., 58 , 17321749.

  • Rotunno, R., , and R. A. Houze Jr., 2007: Lessons on orographic precipitation from the Mesoscale Alpine Programme. Quart. J. Roy. Meteor. Soc., 133 , 811830.

    • Search Google Scholar
    • Export Citation
  • Schultz, D. M., and Coauthors, 2002: Understanding Utah winter storms: The Intermountain Precipitation Experiment. Bull. Amer. Meteor. Soc., 83 , 189210.

    • Search Google Scholar
    • Export Citation
  • Skamarock, W. C., 2006: Positive-definite and monotonic limiters for unrestricted-time-step transport schemes. Mon. Wea. Rev., 134 , 22412250.

    • Search Google Scholar
    • Export Citation
  • Skamarock, W. C., , J. B. Klemp, , J. Dudhia, , D. O. Gill, , D. M. Barker, , W. Wang, , and J. G. Powers, 2005: A description of the Advanced Research WRF, version 2. NCAR Tech. Note NCAR/TN-468+STR, 88 pp. [Available from UCAR Communications, P.O. Box 3000, Boulder, CO 80307].

  • Smith, R. B., , and I. Barstad, 2004: A linear theory of orographic precipitation. J. Atmos. Sci., 61 , 13771391.

  • Smith, R. B., , Q. Jiang, , M. Fearon, , P. Tabary, , M. Dorninger, , and J. Doyle, 2003: Orographic precipitation and air mass transformation: An alpine example. Quart. J. Roy. Meteor. Soc., 129 , 433454.

    • Search Google Scholar
    • Export Citation
  • Stoelinga, M. T., and Coauthors, 2003: Improvement of microphysical parameterization through observational verification experiment. Bull. Amer. Meteor. Soc., 84 , 18071826.

    • Search Google Scholar
    • Export Citation
  • Thompson, G., , R. M. Rasmussen, , and K. Manning, 2004: Explicit forecasts of winter precipitation using an improved bulk microphysics scheme. Part I: Description and sensitivity analysis. Mon. Wea. Rev., 132 , 519542.

    • Search Google Scholar
    • Export Citation
  • White, A. B., , P. J. Neiman, , F. M. Ralph, , D. E. Kingsmill, , and P. O. G. Persson, 2003: Coastal orographic rainfall processes observed by radar during the California land-falling jets experiment. J. Hydrometeor., 4 , 264282.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    (a) WRF analysis of 500-mb geopotential height (solid contours every 60 m), temperature (dashed contours every 5°C), wind barbs (1 full barb = 10 kt) and 300-mb isotachs (shaded light, medium, and dark gray for 50, 60, and 70 m s−1, respectively) from the 36-km WRF at 1200 UTC 4 Dec 2001. Inset boxes denote nested 12-, 4-, and 1.33-km WRF domains. (b) Terrain (shaded, m) and NOAA P-3 flight legs on 4–5 Dec. Dashed lines AB and CD indicate the orientations of the S-Pol vertical cross sections shown in Figs. 9, 10 and 16.

  • View in gallery

    Geostationary Operational Environmental Satellite (GOES) infrared satellite images (K, shaded according to inset scale) at (a) 1200 UTC 4 Dec 2001 and (b) 0000 UTC 5 Dec 2001.

  • View in gallery

    (a) NCEP analysis of 500-mb geopotential height (solid every 60 m), temperature (dashed every 5°C), and winds (1 full barb = 10 kt) at 0000 UTC 5 Dec 2001. (b) As in (a) but for the 12-h WRF 36-km simulation and showing the wind speed at 300-mb (light, medium, and dark gray shading denotes speeds exceeding 50, 60, and 70 m s−1).

  • View in gallery

    (a) NCEP analysis of 850-mb geopotential height (solid every 30 m), temperature (dashed every 5°C), and winds (1 full barb = 10 kt) at 0000 UTC 5 Dec 2001. (b) As in (a) but for the 12-h WRF 36-km simulation. (c) As in (a) but at 1200 UTC 5 Dec. (d) As in (c) but for the 12-h WRF 36-km simulation initialized at 0000 UTC 5 Dec.

  • View in gallery

    (a) Surface observations over western and central Oregon at 0000 UTC 5 Dec (using std surface model, full barb = 10 kt) (b) As in (a) but for the corresponding 4-km WRF simulation showing sea level pressure (solid every 1 mb), surface temperature (dashed every 2°C), and terrain height (shaded every 500 m).

  • View in gallery

    (a) Observed time–height series of equivalent potential temperature (solid) every 1 K using the UW and SLE soundings at (day/time) 04/1200, 04/1630, 05/0000, 05/0130, 05/0330, 05/0600, 05/1200 UTC, as well as winds at 1-h intervals from the IB profiler. (b) As in (a), but for the corresponding 1.33-km WRF simulation output of winds at location IB and equivalent potential temperatures at UW.

  • View in gallery

    Vertical profiles of (a) cross-barrier (west to east) flow (m s−1), (b) potential temperature (short dashed, K) and equivalent potential temperature (solid, K), and (c) the square of moist static stability (N2m) at the UW sounding site at 2100 UTC 4 Dec. Observed (1.33-km WRF-simulated) quantities at the UW site are shown in solid black (solid gray). The 1.33-km WRF moist static stability at point A (Fig. 8b) is shown by the dashed gray line in (c). (d)–(f) As in (a)–(c) but for 0330 UTC 5 Dec.

  • View in gallery

    Observed S-Pol reflectivities (shaded, 1.5° scan) and terrain (solid, 1.0 and 2.0 km MSL contours) at (a) 2000 UTC 4 Dec (c) 0000 UTC 5 Dec, (e) 0400 UTC 5 Dec, and (g) 0800 UTC 5 Dec. The range ring spacing is 50 km. WRF-derived reflectivity (700 mb) from the 1.33-km simulation at (b) 2000 UTC 4 Dec and (d) 0000 UTC, (f) 0400 UTC, and (h) 0800 UTC 5 Dec using gray-shaded scale (every 6 dBZ). The WRF wind vectors at 700 mb are also shown as well as the model terrain (solid, 1.0 and 2.0 m contours). Line AB in (b) is the location for the RHI scans in Figs. 9, 10.

  • View in gallery

    Observed S-Pol reflectivities for cross section AB (shown in Fig. 8b) at 2000 UTC 4 Dec using gray-shaded scale (every 6 dBZ) for (a) 2000 UTC 4 Dec, (c) 0000 UTC 5 Dec, (e) 0400 UTC 5 Dec, and (g) 0800 UTC 5 Dec. The WRF reflectivities, potential temperature (solid every 3 K), and circulation vectors [m s−1 using the inset scale in (b)] from the 1.33-km simulation are shown for (b) 2000 UTC, (d) 0000 UTC, (f) 0400 UTC, and (h) 0800 UTC 5 Dec.

  • View in gallery

    Observed S-Pol reflectivities for cross section AB (shown in Fig. 8b) at (a) 0013 UTC, (b) 0035 UTC, (c) 0056 UTC, (d) 0118 UTC, (e) 0139 UTC, and (f) 0200 UTC 5 Dec using gray-shaded scale (every 6 dBZ). The arrows and numbers indicate the cell locations as they propagate from west to east across the Coast Range and Cascades.

  • View in gallery

    (a) Hovmöller-type plot showing the maximum observed S-Pol radar reflectivity (color shaded every 2.5 dBZ) above 2.0 km MSL from 1900 UTC 4 Dec to 1200 UTC 5 Dec along AB (see Fig. 8b for location). (b) As in (a) but for the 1.33-km WRF.

  • View in gallery

    (a) Observed NOAA P-3 reflectivity (shaded every 6 dBZ) at 1.5 km MSL for legs 2–4 between 2300 UTC 4 Dec and 0100 UTC 5 Dec. There are no data for much of the lee of the Cascades. (b) As in (a) but for the 1.33-km WRF at 0000 UTC 5 Dec as well as the wind vectors at 1.5 km MSL. The boxed region in (a) shows the region included in deriving average cross-barrier wind speeds shown in Fig. 13. The terrain is also shown (0.5 km contours).

  • View in gallery

    (a) Observed wind speeds (shaded every 3 m s−1) and barrier-parallel winds (υ component, dashed every 3 m s−1) from NOAA P-3 Doppler radar averaged over the west–east box across the Cascades for legs 2–4 (boxed region shown in Fig. 12a) between 2351 UTC 4 Dec and 0055 UTC 5 Dec. (b) As in (a) but for the 1.33-km WRF at 0030 UTC 5 Dec. (c) As in (a) but for leg 6 at 0144–0200 UTC 5 Dec. (d) As in (c) but for the 1.33-km WRF at 0200 UTC 5 Dec.

  • View in gallery

    (a) Simulated precipitation (shaded every 5 mm) from the 1.33-km WRF between 2000 UTC 4 Dec to 0800 UTC 5 Dec and terrain (0.5 km contours). (b) Percent of observed precipitation simulated by the 1.33-km WRF at the precipitation gauge sites between 2000 UTC 4 Dec and 0800 UTC 5 Dec. The percent values are color coded using the inset scale in (b), while background shading indicates terrain elevation (m, key at lower left).

  • View in gallery

    Observed vertically pointing radar reflectivity (shaded every 6 dBZ) from the S-Prof (location MB on Fig. 1b) from 1200 UTC 4 Dec to 1200 UTC 5 Dec.

  • View in gallery

    (a) Observed S-Pol radar reflectivity for cross section AB (shown in Fig. 8b) averaged between 0200 UTC and 0600 UTC 5 Dec using gray-shaded scale (every 6 dBZ). (b) S-Pol reflectivity for cross section CD (shown in Fig. 1b) averaged between 2100 UTC 13 Dec and 0100 UTC 14 Dec using gray-shaded scale (every 6 dBZ).

  • View in gallery

    Time series from NOAA P-3, 4-km WRF, and 1.33-km WRF along leg 2 from south (left) to north (right) at 2450 m MSL showing (a) wind speed (m s−1), (b) cloud water (g m−3), vertical velocity (cm s−1), and terrain (m) between 2351 UTC 4 Dec and 0008 UTC 5 Dec.

  • View in gallery

    (a) Observed NOAA P-3 cross section of radar reflectivity (every 2.5 dBZ) and dual-Doppler vertical motions (white contour every 0.3 m s−1) along leg 2 from south (left) to north (right) at 2400 m MSL between 2351 UTC 4 Dec and 0008 UTC 5 Dec. (b) As in (a) but for the 1.33-km WRF simulation output showing circulation vectors in the cross section, cloud water (shaded in g kg−1), snow (solid every 0.1 g kg−1), and graupel (dashed every 0.1 g kg−1). The terrain height within the plane of the section is shown at the bottom of each panel.

  • View in gallery

    Simulated precipitation difference (mm) for the 1.33-km WRF run of (a) SMTH–CTL, (b) NOCR – CTL, and (c) NOCRW – CTL between 2000 UTC 4 Dec and 0800 UTC 5 Dec. The gray shading (every 500 m) in (a) denotes the SMTH terrain, while the shading in (b) and (c) represents the NOCR and NOCRW terrain, respectively. The dashed box in (a) represents the region in which the precipitation was averaged in latitude for Fig. 20.

  • View in gallery

    Simulated precipitation (mm) averaged in latitude over the boxed area in Fig. 19a for the CTL, SMTH, NOCR, and NOCRW runs for the (a) 2000 UTC 4 Dec–0200 UTC 5 Dec and (b) 0200–0600 UTC 5 Dec periods. The terrain profile is indicated at the bottom of the panels. The tables at the top of the panels indicate averages over the coastal range, windward slope, and crest–lee.

  • View in gallery

    Hovmöller plot showing the vertically integrated water and ice (color shaded, mm) for the 1.33-km WRF from 1900 UTC 4 Dec to 0600 UTC 5 Dec along AB (see Fig. 8b for location) for (a) CTL, (b) SMTH, (c) NOCR, and (d) NOCRW experiments. The terrain profile for the CTL run is shown in the lower panel.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 13 13 7
PDF Downloads 15 15 7

Orographic Modification of Convection and Flow Kinematics by the Oregon Coast Range and Cascades during IMPROVE-2

View More View Less
  • 1 School of Marine and Atmospheric Sciences, Stony Brook University, Stony Brook, New York
  • 2 Department of Atmospheric Sciences, University of Washington, Seattle, Washington
© Get Permissions
Full access

Abstract

This paper describes the kinematic and precipitation evolution accompanying the passage of a cold baroclinic trough over the Central Oregon Coast Range and Cascades during 4–5 December 2001 of the second Improvement of Microphysical Parameterization through Observational Verification Experiment (IMPROVE-2) field project. In contrast to previously documented IMPROVE-2 cases, the 4–5 December event featured weaker cross-barrier winds (15–20 m s−1), weaker moist static stability (Nm < 0.006 s−1), and convective cells that preferentially intensified over Oregon’s modest coastal mountain range. These cells propagated eastward and became embedded within the larger orographic precipitation shield over the windward slopes of the Cascades. The Weather Research and Forecasting Model (version 2.2) at 1.33-km grid spacing was able to accurately replicate the observed evolution of the precipitation across western Oregon. As a result of the convective cell development, the precipitation enhancement over the Coast Range (500–1000 m MSL) was nearly as large as that over the Cascades (1500–2000 m MSL). Simulations selectively eliminating the elevated coastal range and differential land–sea friction across the Pacific coastline illustrate that both effects were important in triggering convection and in producing the observed coastal precipitation enhancement. A sensitivity run employing a smoothed representation of the Cascades illustrates that narrow ridges located on that barrier’s windward slope had a relatively small (<5%) impact on embedded convection and overall precipitation amounts there. This is attributed to the relatively weak gravity wave motions and low freezing level, which limited precipitation growth by riming.

Corresponding author address: Dr. Brian A. Colle, School of Marine and Atmospheric Sciences, Stony Brook University, Stony Brook, NY 11794-5000. Email: brian.colle@stonybrook.edu

Abstract

This paper describes the kinematic and precipitation evolution accompanying the passage of a cold baroclinic trough over the Central Oregon Coast Range and Cascades during 4–5 December 2001 of the second Improvement of Microphysical Parameterization through Observational Verification Experiment (IMPROVE-2) field project. In contrast to previously documented IMPROVE-2 cases, the 4–5 December event featured weaker cross-barrier winds (15–20 m s−1), weaker moist static stability (Nm < 0.006 s−1), and convective cells that preferentially intensified over Oregon’s modest coastal mountain range. These cells propagated eastward and became embedded within the larger orographic precipitation shield over the windward slopes of the Cascades. The Weather Research and Forecasting Model (version 2.2) at 1.33-km grid spacing was able to accurately replicate the observed evolution of the precipitation across western Oregon. As a result of the convective cell development, the precipitation enhancement over the Coast Range (500–1000 m MSL) was nearly as large as that over the Cascades (1500–2000 m MSL). Simulations selectively eliminating the elevated coastal range and differential land–sea friction across the Pacific coastline illustrate that both effects were important in triggering convection and in producing the observed coastal precipitation enhancement. A sensitivity run employing a smoothed representation of the Cascades illustrates that narrow ridges located on that barrier’s windward slope had a relatively small (<5%) impact on embedded convection and overall precipitation amounts there. This is attributed to the relatively weak gravity wave motions and low freezing level, which limited precipitation growth by riming.

Corresponding author address: Dr. Brian A. Colle, School of Marine and Atmospheric Sciences, Stony Brook University, Stony Brook, NY 11794-5000. Email: brian.colle@stonybrook.edu

1. Introduction

Several recent field projects, such as the Mesoscale Alpine Programme (MAP; Bougeault et al. 2001), the Intermountain Precipitation Experiment (IPEX; Schultz et al. 2002), and the Improvement of Microphysical Parameterization through Observational Verification Experiment (IMPROVE; Stoelinga et al. 2003) have collected comprehensive kinematic, microphysical, and precipitation datasets over complex terrain. The primary objectives of these field studies were to advance understanding of moist dynamics and cloud microphysics associated with orographic precipitation and to improve bulk microphysical parameterizations (BMPs) in mesoscale models, which will ultimately help improve quantitative precipitation forecasts.

Since orographic precipitation is sensitive to the static stability and terrain variability, this has also been an active area of research. The analysis of recent field datasets has revealed many complex physical mechanisms that can modify orographic precipitation, such as gravity waves in statically stable environments (Bruintjes et al. 1994; Garvert et al. 2007, hereinafter GSM07); development of transient turbulent updrafts observed under statically stable conditions in the presence of strong low-level shear (Houze and Medina 2005; Medina et al. 2005); upstream enhancement in statically stable and blocked-flow conditions (Marwitz 1987a, b; Cox et al. 2005; Colle et al. 2005; Rotunno and Ferretti 2001); and windward enhancement in potentially unstable cases (Browning et al. 1974; Hill et al. 1981; Smith et al. 2003; Medina and Houze 2003), which can organize convection into narrow convective bands (Kirshbaum and Durran 2005).

The importance of gravity waves in modulating orographic precipitation was highlighted by Garvert et al. (2005a; GSM07) using aircraft in situ and radar observations as well as high-resolution (1.33-km grid spacing) simulations from the fifth-generation Pennsylvania State University–National Center for Atmospheric Research (NCAR) Mesoscale Model (MM5). They showed enhancement of vertical motions and precipitation over the narrow ridges of the windward (west facing) slopes of the Oregon Cascades during the 13–14 December 2001 IMPROVE-2 heavy rainfall event, which resulted in maxima in cloud water and snow production extending up to 6 km MSL over these ridges. When the low-level flow is blocked, the orographic precipitation is shifted upstream of the windward slope (Colle et al. 2005; Rotunno and Ferretti 2001; Bousquet and Smull 2003b), and there are fewer gravity wave and precipitation perturbations over the windward ridges (Colle 2008).

Recent field studies have also added insight into the details of precipitation evolution during the passage of baroclinic waves over complex terrain. Ground-based radar from MAP revealed that in some cases the orographic precipitation enhancement was relatively steady over the windward slope but occasionally there were embedded convective cells of precipitation that propagated up over the barrier and locally intensified over the steep Alpine ridges (Smith et al. 2003). Using vertically pointing radar data from IMPROVE-2, Medina et al. (2007) developed a conceptual model of the commonly observed progression of precipitation features as extratropical cyclones traverse mountainous terrain in the Pacific Northwest. As a cyclone approaches, the precipitation echo first appears aloft and descends toward the surface [leading edge echo (LEE)]. During the time of maximum large-scale ascent, the precipitation consists of a vertically continuous layer extending from the windward slope up to 5–6 km MSL that persists for several hours. During this period, there is often a double maximum echo (DME) over the windward slopes. One maximum is associated with the melting-induced bright band. A secondary maximum (1–2.5 km above the bright band) results from or is enhanced by interaction of the baroclinic system with the terrain, according to Medina et al. (2007). Finally, the shallow convection echo (SCE) stage is associated with low echo tops in the more unstably stratified later portion of the storm, which is typically postfrontal.

A variety of synoptic environments were observed during IMPROVE-2. Most of the detailed orographic precipitation and microphysical studies conducted thus far have focused on the strong cross-barrier flow events with well-defined frontal passages under moist stable conditions, such as 13–14 December 2001 and 27–28 November 2001 (Garvert et al. 2005a; Medina et al. 2007). These cases involved a deep orographic cloud over the Cascades, with little transient convective activity propagating over the Cascades. These events need to be compared with storms featuring weaker cross-barrier flow and stability, since these parameters impact terrain-induced gravity wave contributions to orographic precipitation (Colle 2004; Smith and Barstad 2004) as well as the stratiform versus convective nature of the precipitation (Rotunno and Houze 2007). There have been no formal IMPROVE-2 studies highlighting the evolution of intermittent convection from the Coast Range to the Cascades, so this study focuses on the 4–5 December event, which had a reduced cross-barrier flow, weaker stability, and lower freezing level than previously documented IMPROVE-2 cases.

Most previous simulation studies for IMPROVE-2, IPEX, and MAP events have utilized MM5. The Weather Research and Forecasting Model (WRF; Skamarock et al. 2005), which is becoming more widely used, includes higher-order numerics and new BMPs [such as that developed by Thompson et al. (2004)]. There have been few formal studies of orographic precipitation using WRF, so this study provides an opportunity to evaluate this relatively new model using the unique high-resolution IMPROVE-2 dataset.

The objectives of this paper are to document the passage of a baroclinic wave accompanied by embedded convection over the Coast Range and Cascade Mountains of western Oregon in order to 1) describe the flow kinematics, precipitation evolution, and orographic modification and compare with other IMPROVE-2 events; 2) verify the WRF-simulated mesoscale flow and precipitation patterns using comprehensive in situ and aircraft datasets; and 3) highlight the intermittent nature of the convective precipitation and isolate the respective impacts of the Oregon Coast Range and narrow Cascade ridges on precipitation over this complex region. The case analyzed herein is the 4–5 December 2001 IMPROVE-2 event, corresponding to intensive observing period (IOP) 6. A subsequent paper will highlight the microphysical observations and verification of the WRF BMPs for this event.

Section 2 provides a detailed description of observational data and model configurations. Section 3 presents the synoptic evolution of the event and comparison with the model. Section 4 focuses on the kinematic evolution and the associated orographic precipitation response. Section 5 includes a comparison of these results with the previously documented IMPROVE events. Summary and conclusions are presented in section 6.

2. Data and methods

a. Observational datasets

The IMPROVE-2 project collected a comprehensive set of observations to evaluate the thermodynamic and kinematic structures as well as cloud microphysics of winter storms (Stoelinga et al. 2003). The primary observational facilities during the 4–5 December event used in this study are indicated in Fig. 1b. The NCAR S-band dual-polarization Doppler radar (S-Pol), which was situated in the lower western foothills of the Cascade Mountains, provided a three-dimensional view of the precipitation structures extending from the adjacent Willamette Valley eastward up to the Cascade crest. The University of Washington (UW) mobile sounding unit and NCAR integrated sounding system (ISS) at Irish Bend (IB) were used to obtain the winds and stratification just upstream of the Cascades. The National Oceanic and Atmospheric Administration/Earth System Research Laboratory (NOAA/ESRL) vertically pointing S-band radar (S-Prof) was deployed on the windward slope of the Oregon Cascades at ∼500 m MSL (see MB on Fig. 1b). Specifications for S-Prof are provided by White et al. (2003), while Stoelinga et al. (2003) and Medina et al. (2007) describe its operation in IMPROVE-2.

The NOAA P-3 and UW Convair aircraft collected in situ cloud microphysical data as well as the ambient wind, temperature, and moisture data over the Cascades. Since there were limited observations by the Convair for this IOP, this study utilizes the data from the NOAA P-3, which executed multiple flight legs in order to provide continuous, high-resolution dual-Doppler radar coverage (Fig. 1b). As described in Bousquet and Smull (2003a), volumetric radial velocity and reflectivity measurements were edited to remove ground clutter contamination, noise, and other artifacts. The radial velocity data were then interpolated to a composite Cartesian grid (240 km × 170 km) with 1-km grid spacing in the horizontal and 0.25-km resolution in the vertical up to 10 km MSL (or echo top, if lower). Radial velocities from the P-3 tail radar were synthesized to estimate 3D airflow over the Cascades as in GSM07.

b. Model setup

The WRF, version 2.2 (v2.2), was utilized to simulate the 4–5 December 2001 IMPROVE-2 event. The 36-, 12-, and 4-km domains are the same as those used in Garvert et al. (2005a; GSM07) for the 13–14 December event (Fig. 1a), while the inner 1.33-km domain was extended farther westward to include the Oregon Coast Range and adjacent Pacific waters. Two simulations were completed using initial and time-dependent boundary conditions from the National Centers for Environmental Prediction (NCEP) Global Forecast System (GFS) 6-h analyses. The first 18-h WRF simulation was initialized at 1200 UTC 4 December 2001, while another 12-h WRF run was started at 0000 UTC 5 December. Two separate control simulations were needed, since the first (earlier) simulation had too much subsidence drying with the midlevel trough after 0600 UTC 5 December and thus too little precipitation. Thirty-two unevenly spaced half-sigma levels were used in the vertical, with maximum resolution in the boundary layer. The control simulation used the updated Reisner2 scheme (Thompson et al. 2004), updated Kain–Fritsch cumulus parameterization (Kain 2004), and Eta Mellor–Yamada–Janjić (MYJ) PBL (Janjić 1994), which are similar to those physics used in Garvert et al. (2005a). The convective parameterization was turned off for the 4- and 1.33-km domains. All simulations used the positive definite advective (PDA) scheme in WRF v2.2 for the moisture and hydrometeor advection (Skamarock 2006). As compared with an equivalent WRF run without PDA, the PDA resulted in a large (20%–30%) reduction of precipitation in the more convective areas near the coast at 1.33-km grid spacing. The non-PDA advective scheme has been shown to generate artificial water mass after setting negative mixing ratios to zero (Braun 2006).

3. Synoptic-scale evolution

a. Upper-level and surface analysis

The 4–5 December event featured two short-wave troughs rotating around a broad upper-level trough centered over the Pacific Northwest. At 1200 UTC 4 December 2001 (Fig. 1a), the first 500-mb (hPa) short wave crossed over the IOP area. Meanwhile, another short-wave trough was located around 140°W, in association with a 60 m s−1 jet at 300 mb (Fig. 1a) and an area of mid- and upper-level clouds around 47°N, 130°W (Fig. 2a). The open-cellular convection to the north and east of this offshore cloud shield at this time suggests that this baroclinic development was occurring in a relatively cold and unstable air mass left in the wake of the first short-wave trough. At 850 mb (not shown), there was a well-defined trough over eastern Washington and Oregon, with west-northwesterly flow of ∼15 m s−1 over the study region.

As the offshore 500-mb trough approached the Pacific Northwest coast at 0000 UTC 5 December in the NCEP analysis and the 12-h WRF forecast (Figs. 3a,b), the mid- and upper-level clouds increased their spatial coverage, and cloud tops cooled to ∼220 K near the Oregon coast (Fig. 2b). This development occurred in a favored region for upward motion in the downstream exit region of a 70 m s−1 jet along the west side of the advancing trough (Fig. 3b). The WRF realistically predicted the 850-mb trough that extended southward along the southwest British Columbia, Washington, and Oregon coasts at this time (Figs. 4a,b), where winds shifted from west-southwesterly over land to west-northwesterly a few hundred kilometers offshore. The north–south trough orientation is similar to that found in the 13–14 December and 28–29 November IMPROVE-2 storms examined by Medina et al. (2007); however, the trough observed at 0000 UTC 5 December was not as sharp and was accompanied by only weak warm advection. Relative to these other two IMPROVE-2 events, the 850-mb winds of 12–15 m s−1 at 0000 UTC 5 December were less than half as strong, while the 850-mb temperatures over the Cascades (−4°C) were about 10°C cooler.

Across western and central Oregon at 0000 UTC 5 December, both surface observations and the 4-km WRF run showed 5–10 m s−1 southwesterly winds at the coast that were deflected northward by the Cascades and thus flowed down the pressure gradient within the Willamette Valley (Fig. 5). The wind directions suggest that some of this air had previously crossed over the coastal terrain and merged with channeled flow arriving from the south.

By 1200 UTC 5 December, the 500-mb trough axis had advanced into western Washington and Oregon (not shown). The associated 850-mb trough was located over eastern Oregon at this time (Figs. 4c,d), and weak cold advection and west-northwesterly flow covered the study region. The 12-h WRF forecast realistically captures the trough’s position and strength of the cross-barrier flow at this time.

b. Vertical structure

A time–height section of winds from the 915-MHz wind profiler at Irish Bend (IB in Fig. 1) and equivalent potential temperatures (θe) derived from the UW and Salem (SLE) soundings illustrates the observed wind and temperature profiles over western Oregon during the event (Fig. 6a) and may be compared with those depicted by the 1.33-km WRF simulation (Fig. 6b). Early in the IOP (1800 UTC 4 December), there were relatively weak (<10 m s−1) south-southwesterlies below 1 km MSL associated with the terrain blocked and channeled flow within the Willamette Valley, which veered to westerly at 7–10 m s−1 above 1.5 km MSL. By 0000 UTC 5 December, the low-level southwesterlies deepened to 3 km MSL and increased to 15–20 m s−1 between 1 and 3 km MSL as the upper-level short-wave trough approached. Meanwhile, the veering from 2 to 4 km MSL between 2100 UTC 4 December and 0300 UTC 5 December is indicative of weak warm advection. After 0300 UTC 5 December, the observed stratification was weaker above 5 km MSL, where e/dz ∼ 0. The model simulated the increased southwest winds and warm advection from 2230 UTC 4 December to 0330 UTC 5 December 2001 (during the period of the P-3 flight) as well as the decrease in stability aloft.

A forward-tilted trough extending from the surface to 3 km MSL crossed the IB site at 0600 UTC 5 December at 3 km and at the surface at 0900 UTC 5 December (Fig. 6a). A reduction in stability accompanied the trough’s passage around 3–4 km MSL, with equivalent potential temperatures decreasing after trough passage. The simulated trough-induced wind shift at IB was weaker than observed, especially between 2 and 3 km MSL. During the trough passage, the simulated θe values were higher than those observed and the simulated stratification at low levels was weaker than what was observed. However, there were no sounding data collected for this period (0600–1200 UTC 5 December), which limits the extent to which the model can be validated. The model produced too little drying in the lowest 2 km after trough passage, so simulated θe’s were 3–4 K greater than observed after 0900 UTC 5 December.

The UW sounding was used to quantify the stratification and the vertical wind profile just upwind of the Cascades (Fig. 1b). At 2100 UTC 4 December, the observed cross-barrier (westerly component) flow increased from near −3 m s−1 at the surface to +15 m s−1 at 1.75 km MSL (black line in Fig. 7a). The WRF underpredicted the cross-barrier flow at the 1.5 km level and hence underestimated the strength of the low-level shear (gray line in Fig. 7a). The WRF-simulated potential temperature profile was close to what was observed (Fig. 7b), but the θe values in the WRF simulation were 1–2 K greater than observed (Fig. 7b), suggesting that the model was slightly too moist. The square of the moist Brunt–Väisälä frequency (N2m), as defined in Durran and Klemp (1982), was less than zero just below 1.5 km MSL in both the model and observations (solid gray and black lines in Fig. 7c, respectively). Both the observed and simulated atmospheres were subsaturated below 700 m (not shown), indicating stable stratification at these levels (dry N of ∼0.005 s−1), which favored some near-surface flow blocking (Fr = U/Nhm ∼ 0.8, where hm is 2.0 km and U of 8 m s−1) as previously noted in Fig. 5. Meanwhile, between 0.7 and 2.0 km, the 10–15 m s−1 cross-barrier flow (Fig. 7a) was saturated with weak stratification (Fig. 7c), so there was little potential for flow blocking at these levels.

Although there were no soundings available at the Oregon coast, Fig. 7c shows the WRF-simulated N2m profile at point A (near the coast, Fig. 8b) at 2100 UTC 4 December (dashed gray line in Fig. 7c). The air over the coast was saturated above 0.7 km MSL and in the lowest 2 km MSL it was potentially unstable [N2m < 0 in Fig. 7c, with maximum CAPE ∼200 J kg−1 (not shown)].

By 0330 UTC 5 December, the moist static stability in the observed UW sounding had increased to 0.02 s−1 around 1.0 km MSL (Fig. 7f), which coincided with the freezing level. This is suggestive of additional cooling from melting of snow and evaporation below this level. The model failed to reproduce this stable layer, in part because the simulated PBL was too moist (saturated) below 1.0 km, as revealed by the θe profile (Fig. 7e). The WRF realistically predicted the winds (Fig. 7d), although it had a somewhat greater cross-barrier component below 1 km, which is indicative of underestimated near-surface blocking in the model. The average moist Froude number below 1.0 km at the UW site was ∼0.75 and 1.0 in the observed and WRF-simulated profiles, respectively, based upon respective average Nm values of 0.008 s−1 and 0.006 s−1 and similar layer-averaged U values near 10 m s−1 (Figs. 7d,f). Meanwhile, above 1.0 km the mean stability was slightly less (Nm ∼ 0.005 s−1), the flow was stronger (U ∼ 15 m s−1), and parcels only had to surmount the remaining ∼1.0 km of the Cascades. Thus the moist Froude number was ∼3.0, which suggests little if any flow blocking above midmountain level. At this time, the layer of potential instability below 2 km MSL observed earlier at the coast (point A in Fig. 8b) still persisted (dashed gray line in Fig. 7f).

4. Mesoscale precipitation and kinematics

a. Precipitation evolution

Figure 8 shows representative 1.5° elevation scans of S-Pol radar reflectivity and 1.33-km WRF-simulated reflectivity1 at 700 mb, which highlights the change in the precipitation structures during the event. Because of the uncertainties in obtaining simulated reflectivity, only qualitative comparisons are made between observed and simulated reflectivity structures. Early in the IOP at 2000 UTC 4 December (Figs. 8a,b), convective cells were observed over the Cascades and Coast Range, consistent with the prevailing low-level potential instability at this time. The model had less instability (Fig. 7c) and hence reduced convective activity as compared to the observations. By 0000 UTC 5 December (Figs. 8c,d), the observed precipitation became more widespread and stratiform over the Cascades. By 0400 UTC 5 December, the time preceding the midlevel trough passage (Fig. 6), the precipitation coverage increased and reached its maximum extent around the radar (Figs. 8e,f). The precipitation was mainly stratiform (<30 dBZ) with some embedded convective cells over the coastal range and around the radar site. By 0800 UTC 5 December, a time coinciding with the passage of the midlevel trough (Fig. 6), the precipitation had become more cellular to the north of the S-Pol site (Figs. 8g,h). As shown in Fig. 8, the model simulation realistically captured the overall evolution of the precipitation.

A series of west-to-east cross sections (denoted AB in Fig. 8b) illustrate the evolving precipitation pattern over the Cascades and Coast Range (Fig. 9). At 2000 UTC 4 December (Figs. 9a,b), both the radar and 1.33-km WRF simulation depict narrow convective plumes below 4 km MSL within the broader precipitation shield blanketing the Coast Range and Cascades. According to the 1.33-km WRF simulation (not shown), some of these convective cells originated well offshore over the eastern Pacific, which is consistent with the open-cellular convection in the satellite imagery early in the event (Fig. 2a). The cells intensified as the airflow ascended over the Coast Range (Figs. 9b,d), releasing potential instability at low levels. Meanwhile, reflectivity values decreased sharply toward the lee of the Cascades as a result of leeside subsidence accompanying a moderate standing wave anchored to the crest (Fig. 9b).

The precipitation coverage along cross section AB increased by 0000 UTC 5 December (Figs. 9c,d). Some of the strongest echoes at this time were over the Willamette Valley near the S-Pol site. Both observed and simulated reflectivity values and echo coverage continued to increase over the windward Cascade slopes until 0400 UTC 5 December (Figs. 9e,f). At this time the midlevel reflectivity contours (12–18 dBZ) sloped upward so as to parallel the underlying terrain. Meanwhile, narrow convective plumes had developed 25–50 km to the west of the S-Pol between 5 and 7 km MSL in conjunction with decreasing stability in this layer with the approach of the upper-level trough (Fig. 6a). The simulated stability was also weak in this layer (Fig. 6b), but the model did not generate any corresponding elevated convection. Precipitation spillover east of the Cascade crest had increased in comparison to a few hours earlier, with more spillover suggested in the WRF simulation than the observations. This trend continued until 0800 UTC 5 December (Figs. 9g,h), with substantially more precipitation (including embedded convective cells) extending into the lee of the Cascades as the upper-level trough crossed the barrier and moved over eastern Oregon.

To illustrate further the transient and convective nature of the precipitation during this event, Fig. 10 shows six observed S-Pol RHI scans along AB every ∼22 min from 0013 to 0200 UTC 5 December. At 0013 UTC (Fig. 10a), there was an intense precipitation cell (>24 dBZ) over the Willamette Valley (cell number 1), while there was broader area of enhanced precipitation over the Cascade windward slope. Cell number 1 moved eastward such that by 0139 UTC it was over the Cascade crest (Fig. 10e), and by 0200 UTC it dissipated in the lee (Fig. 10f). Meanwhile, cell number 2 moved onshore between 0013 and 0056 and became enhanced over the Coast Range (Figs. 10a–d), and subsequently weakened over the Willamette Valley (Figs. 10e,f). Cell number 3 made landfall and became enhanced over the Coast Range between 0118 and 0200 UTC 5 December (Figs. 10d–f), while relatively steady enhanced precipitation was observed over the Cascade windward slope.

Figure 11 shows Hovmöller-type plots (i.e., as a function of time and distance from the radar along section AB from 1900 UTC 4 December to 1200 UTC 5 December 2001) comparing maximum reflectivity values above 2.0 km MSL as derived from S-Pol observations and the 1.33-km WRF simulation. This level was chosen since it is high enough to exclude brightband contamination near the melting level yet low enough to capture the essential pattern of orographically enhanced precipitation.2 During the first ∼7 h of the event, both the S-Pol and WRF showed areas of slightly enhanced reflectivity over the Pacific Ocean that further intensified as they reached the coastal range. Beginning around 1900–2100 UTC 4 December, scattered convection (both observed and simulated) increased in coverage over the windward slopes of the Coast Range and Cascades, with little spillover beyond the Cascade crest. Cells developed over the Coast Range and propagated eastward over the Cascades,3 weakening as they passed over the east side of the Coast Range as a result of subsidence in the lee of this modest barrier (Fig. 9). The WRF simulation produced more cells over the coastal range than what was observed (Figs. 8c,d); thus, it had more cell streaks on the Hovmöller plot. This suggests that the WRF simulation may have been slightly more unstable than observed in the coastal region.

After 0000 UTC 5 December there was a gradual increase in the amount of precipitation spilling over the Cascade crest. With the trough’s approach between 0200 UTC and 0700 UTC 5 December, the precipitation intensified over the Cascades and became more widespread in the S-Pol observations (Fig. 11), while new clusters of cells continued to be enhanced over the Coast Range and then advanced eastward over the Cascades. The maximum reflectivities in the WRF simulation extended to the second peak near the crest (at ∼80 km in Fig. 11b), which was less pronounced in the S-Pol, suggesting that the model may have produced too much precipitation and spillover beyond the crest. By 0700 UTC 5 December, the trough had passed over the Willamette Valley at around 4 km MSL (Fig. 6), and the precipitation weakened over the Coast Range and Cascades. Following passage of the frontal rainband over the S-Pol site at 0900 UTC 5 December, the precipitation became more scattered and convective in both the observations and simulation.

The reflectivity sequence seen during the 4–5 December case as illustrated in Fig. 11 has some similarities to the convective cells that developed during MAP IOP2b (Asencio et al. 2003; Smith et al. 2003; Georgis et al. 2003; Rotunno and Houze 2007) as southerly and statically unstable airflow ascended the Apennines, a coastal barrier south of the Alps. In the lee of the Apennines, subsidence weakened the cells, which then drifted northward and became embedded in the stratiform, orographically forced cloud over the Alps and increased their intensity over local peaks (Asencio et al. 2003; Medina and Houze 2003; Smith et al. 2003).

b. NOAA P-3 observations

The simulated precipitation structures and airflow over the Cascades were compared with the winds derived from the NOAA P-3’s Doppler radar data obtained from legs 1–4 between 2300 and 0030 UTC 5 December (Fig. 1b). At 1.5 km MSL, the local maxima of reflectivity were not clearly collocated with the individual ridges at this time (Figs. 12a,b). The observed winds at 1.5 km MSL decelerated from 15–20 m s−1 to 5–10 m s−1 and became more southwesterly adjacent to the crest. The 1.33-km WRF-simulated winds were roughly 2–5 m s−1 stronger than the observations upstream of the Cascades. At 2.5 km MSL (not shown), fairly uniform west-southwesterly flow ∼20 m s−1 prevailed across the Cascades in both the P-3 observations and the WRF simulation. At this level the maximum precipitation tended to be more focused over the narrow windward ridges.

Figures 13a,b shows the cross- and along-barrier winds averaged over a portion of the Cascades as derived from the P-3 Doppler radar from 2330–0100 UTC and the 1.33-km WRF at 0030 UTC. The higher-momentum air (>18 m s−1) observed by the NOAA P-3 sloped upward over the windward slopes of the Cascades between 2 and 3 km MSL. There was a well-defined low-level shear layer, which has been shown to be a robust feature of orographic winter storms (Houze and Medina 2005; Medina et al. 2005). The WRF simulation developed a wind speed maximum over the slope similar to the observations, but the simulated shear layer was both too shallow and too weak—a shortcoming that was also noted in MM5 simulations conducted by GSM07. Alternate PBL schemes tested within the WRF model framework exhibit the same problem, suggesting that further attention is needed to properly represent PBL processes leading to this terrain-induced low-level shear layer.

As the P-3 flew westward across the Cascades at 0200 UTC along leg 6 (Fig. 1b), winds accelerated from 18 m s−1 over the lower windward slope to 24–27 m s−1 over the crest at 3–4 km MSL (Fig. 13c). The WRF-simulated flow maximum over the crest was weaker than the observations. The WRF-simulated airflow suggests that there was a mountain wave over the barrier.

c. Surface precipitation distribution

Figure 14a shows the total simulated precipitation from the WRF 1.33-km domain from 2000 UTC 4 December to 0800 UTC 5 December. The heaviest simulated precipitation (45–55 mm) occurred ∼45.2°N over the Coast Range and ∼44.9°N over the lower windward slope of the Cascades. Interestingly, the precipitation amounts over the Coast Range were nearly the same as over the far taller and wider Cascades. In the IOP region of the Cascades (boxed region in Fig. 12a), the simulated precipitation over the windward slope ranged from 35–45 mm over the ridges to 15–20 mm in the major valleys. There was a rapid reduction of precipitation to less than 20 mm in the immediate lee of the Cascade crest.

Figure 14b shows the percent of observed precipitation that was simulated by the 1.33-km WRF at the Snowpack Telemetry (SNOTEL) and hourly cooperative observer sites from 2000 UTC 4 December to 0800 UTC 5 December. Given a characteristic tendency for undercatchment of snow, model overprediction is typically not diagnosed until this ratio exceeds 150% (Colle et al. 2000; Garvert et al. 2005b). Most values in the present case ranged from 100%–130% over the central Cascades, indicating no tendency for widespread model overprediction. There is some suggestion of model overprediction immediately downwind of the Cascades, while farther to the east (over the eastern Oregon plateau) the WRF underpredicted precipitation at many sites by at least 30%. The model also underpredicted precipitation by 35%–50% at a few coastal observation sites and Willamette Valley, and this error extended northward within the larger 4-km domain (not shown). A subsequent paper will relate this precipitation verification to the microphysical aircraft data aloft.

5. Discussion

a. High-resolution reflectivity evolution

The S-Prof radar illustrates how the precipitation structures changed at a particular point (MB in Fig. 1a) over the windward Cascade slopes. The S-Prof observations (Fig. 15) are compared with the conceptual model of Medina et al. (2007), which was developed using the most typical and stronger IMPROVE-2 IOPs (i.e., excluding 4–5 December). They noted the initial decent of precipitation originating well aloft as cyclones approached the radar site, which they referred to as the LEE pattern. Warm advection was relatively weak in the leading sector of the 4–5 December baroclinic wave (Fig. 4); thus, there was less potential for gradual ascent and a lowering cloud deck. In addition, the baroclinic wave of interest passed over the area only 6–7 h after passage of the previous frontal wave, yielding an atypical initial period of reduced stability. Hence precipitation began in the form of relatively shallow convective cells rather than a thickening (and lowering) stratiform shield. Between 1600 and 1900 UTC 4 December (Fig. 15), there were periods of convective precipitation extending up to 3 km MSL as the convective cells traversed S-Prof’s location; these cells became more frequent and deeper by 2000 UTC 4 December.

By 0000 UTC 5 December, the orographic cloud had deepened to 5.5 km MSL. A region of large vertical gradients of reflectivity was located at ∼3 km MSL, near the layer of most rapid model-predicted depositional growth (around −15°C). Embedded convective cells appeared above 4 km MSL after 0400 UTC 5 December as the layer of enhanced reflectivities at ∼3 km MSL progressively thinned so as to resemble the upper member of the DME described by Medina et al. (2007). The lower member, marking the bright band, was not observed given the low (<1.0 km MSL) prevailing 0°C level.

After 0800 UTC 5 December, there were more narrow plumes of convective enhancement between 3 and 4 km MSL. An SCE (Medina et al. 2007) stage associated with lower echo tops and more convective precipitation marked the last stage of the passing storm (after 1100 UTC 5 December).

The results of GSM07 and Medina et al. (2007) suggest that the reflectivity enhancement above the Cascade crest around the time of maximum large-scale ascent may be associated with a mountain-wave circulation aloft. For example, Fig. 16b shows a time-mean vertical cross section of S-Pol reflectivity along line CD (Fig. 1b), corresponding to the previously analyzed case of 13–14 December 2001, averaged during the period highlighted in GSM07 (2300 UTC 13 December to 0100 UTC 14 December). This cross section can be compared to a similar one along line AB (Fig. 1b) averaged from 0200 to 0600 UTC 5 December 2001 (Fig. 16a). On 13–14 December 2001 precipitation was deeper (>8 km MSL), with a pronounced bright band at 1.5 km MSL. Over the upper windward Cascade slopes the reflectivity bulged upward to 6 km MSL in response to a larger-scale mountain wave anchored to the crest (GSM07). In contrast, precipitation on 5 December was shallower (<6 km MSL). The 0°C level at this time was very low (<1 km) and therefore a well-defined bright band could not be observed. For this case, the upward bulge of the reflectivity over the crest was very subtle. This is consistent with the comparatively weak mountain wave over the crest on 5 December, for which simulated upward motions were <0.5 m s−1 above 5.0 km MSL (Fig. 9f) as compared with 0.5–1.0 m s−1 in the presence of far stronger cross-barrier flow on 13–14 December (cf. Fig. 8 of Garvert et al.).

b. Gravity wave impacts on precipitation

The WRF-simulated cross sections suggest that the vertical motions depicted over the Cascades were associated with gravity waves over the small-scale (10–20-km wide) ridges, especially toward the middle of the event as stability increased (Fig. 9f). Figure 17 shows P-3 flight-level data and WRF output for the north–south leg 2 at 2.4 km MSL between 2351 UTC 4 December and 0008 UTC 5 December. The WRF underpredicted the flow by 3–5 m s−1 on average (Fig. 17a), which was also apparent upstream at UW near the 2.4-km altitude (Fig. 7a). As in other IMPROVE-2 IOPs (GSM07), the WRF was able to realistically simulate the vertical velocity perturbations of ±1 m s−1 over the ridges as the southwesterly cross-barrier flow of 15–20 m s−1 produced small-scale gravity waves over the narrow ridges (Fig. 17c). At 1.33-km resolution, the simulated vertical velocity maxima were underpredicted by 30%–40% given the weaker-than-observed flow. In contrast, the 4-km WRF could not reproduce these small-scale oscillations given its inherently smoother topography. Both observations and simulation results in Fig. 17 indicate that vertical velocities over the windward slopes were only about half as large as those on 13–14 December 2001 at this altitude (GSM07). As such, observed supercooled water variations were not as well correlated with areas of observed upward motion as for the 13–14 December event. On 4–5 December, WRF-simulated cloud water variations over the ridges were more pronounced than shown by in situ observations (Fig. 17b), suggesting a persistent deficiency in the model microphysics that will be explored in a subsequent paper.

An outstanding question is whether these wave perturbations were still large enough to impact the precipitation over the Cascades. Figure 18 shows the observed NOAA P-3 reflectivity field within a vertical section along leg 2 and the corresponding 1.33-km simulated snow, graupel, and cloud water mixing ratios. In the observations (Fig. 18a), little correspondence is noted between the areas of heaviest precipitation (x = 47 and 110 km) and the narrow ridges, as conditions were influenced significantly by translating convective cells originating westward (Fig. 11). This is in sharp contrast to conditions on 13–14 December depicted for this same general “leg 2” location (cf. Fig. 14 of GSM07), in which well-defined and locally forced plumes of enhanced reflectivity (>18 dBZ) persisted above each of these ridges. However, the simulation does depict some correspondence between the enhanced cloud liquid water and the individual ridges, which will be explored in a follow-up paper on the microphysics. The amount of simulated graupel and supercooled water over the ridges in Fig. 18b is about 50% less than on 13–14 December (cf. Fig. 15 of GSM07), which likely further limited the ridgeline precipitation in the 4–5 December case.

GSM07 quantified the impact of the narrow windward ridges on the net precipitation over the Cascades at 1.33-km grid spacing by completing a run in which the terrain was smoothed so as to be similar to that employed in the 12-km WRF domain. We performed the same smoothed (SMTH) terrain experiment as described in GSM07, while maintaining identical physics and upstream terrain for the Coast Range as in the control run. Figure 19a depicts the smoothed Cascade terrain and resulting spatial precipitation pattern of precipitation differences between the SMTH and control (CTL) 1.33-km runs for the 2000 UTC 4 December to 0800 UTC 5 December period. The CTL run produced 5–7 mm (5%–12%) more precipitation over some of the windward ridges and 4–6 mm (10%–20%) less precipitation in the east–west-oriented valleys as compared to the SMTH run (Fig. 19a). Relative to the SMTH run, the CTL run produced 9–12 mm (10%–25%) more precipitation over the crest given the adjacent steep windward slope, and it had 1–2 mm (10%–20%) less precipitation in the Cascades lee given the water vapor depletion over the Cascades (Fig. 20).

Figure 20 shows the distribution of simulated precipitation in the cross-barrier direction averaged in latitude over the dashed region in Fig. 19a between 2000 UTC 4 December and 0200 UTC 5 December and 0200 and 0600 UTC 5 December for several sensitivity runs. Unlike GSM07, who noted a 10%–15% increase in net precipitation on 13–14 December tied to realistic small-scale corrugations of Cascade windward slope, only a 3%–6% net increase was found between the CTL and SMTH runs for the 4–5 December event. A Hovmöller plot of the vertically integrated water and ice for the SMTH run reveals only minor differences in precipitation evolution over the Cascades as compared to the CTL run (Figs. 21a,b), which suggests that these narrow ridges along the windward slopes were not particularly important in modulating the development or evolution of the convective precipitation in this case. Although the 1.33-km WRF simulation underpredicted the small-scale vertical motions over the windward slope, the impact that the narrow ridges had on the actual precipitation was likely small given the inconsistent phasing between observed reflectivity maximum and individual ridges (Fig. 18a).

Overall, the reduced impact of small-scale terrain features in the Cascades in our case compared to that of GSM07 is attributed to two factors. First, as highlighted above, the gravity wave vertical motions were about half as strong as in GSM07, so there was less supercooled water generation and riming (which have been shown to increase the precipitation efficiency over windward ridges; Medina and Houze 2003; Colle and Zeng 2004). Secondly, a relatively low freezing level on 4–5 December, (∼1000 m, as compared with ∼2000 m for 13–14 December) suggests less supercooled water and riming and hence less local precipitation fallout over individual ridges. This is consistent with Colle (2008), who used idealized two-dimensional MM5 simulations to show that the net precipitation enhancement over a series of 8–10 narrow ridges increased as the freezing level was raised from 1000 to 750 mb.

c. Coast Range impacts

GSM07 illustrated that the interaction of onshore flow with the coastal range can reduce the cloud water in the lee of this modest barrier (viz. over the Willamette Valley), but did not quantify the associated surface precipitation impacts. We completed a sensitivity experiment in which the coastal range within the 1.33-km WRF domain was replaced by a flat land surface (NOCR). With less upward motion to release the potential instability over the coastal range (not shown), the precipitation over the “coastal strip” (i.e., the inland zone extending ∼50 km east from the Oregon coast) was reduced by 40%–60% (Figs. 19b, 20). Meanwhile, with more moisture available downstream of the coastal strip, the precipitation over the Willamette Valley and lower windward slopes of the Cascades correspondingly increased by 10%–20% in the NOCR simulation. Cellular convection was slightly weaker over the coastal strip in NOCR than in CTL (Figs. 21a,c), but slightly greater over the Willamette Valley and the lower windward slopes of the Cascades.

Interestingly, even after removing the elevated coastal-range terrain, a zone of enhanced precipitation persisted over this region and convective cells continue to intensify shortly after landfall (Fig. 21c and dotted lines in Fig. 20). This suggests that differential friction at the coast might also have played a role in producing low-level convergence and upward motion at the coast, which released the potential instability and enhanced the coastal precipitation. The importance of differential friction in enhancing coastal precipitation has been highlighted in other studies (Braun et al. 1999; Colle and Yuter 2007). In the 1.33-km CTL WRF simulation at 0000 UTC 5 December (Fig. 5b), the surface convergence was ∼4 × 10−4 s−1 along the coast (not shown), which resulted in enhanced vertical motion and release of the potential instability. To illustrate the importance of the differential friction, an additional 1.33-km simulation was completed in which the area of the (already flattened) coastal mountains was replaced with a water surface (NOCRW). This change reduced the surface precipitation by 60%–70% in this zone as compared to the CTL run (Figs. 19c, 20). Moreover, convective cell coverage and intensity were reduced both over the coastal strip and adjacent waters of the Pacific (Fig. 21d). Meanwhile, more intense precipitation cells were triggered over the Willamette and the lower reaches of the Cascades. This illustrates that the precipitation maximum over the Coast Range, whose amplitude was nearly equivalent to that over the Cascades in the CTL run (Fig. 20), resulted from a superposition of two effects: 1) lifting induced by the presence of an elevated terrain profile corresponding to the coastal mountains, and 2) differential land–sea friction and associated low-level convergent forcing tied to the Pacific coastline.

Accumulated precipitation over the zone extending from the base of the windward Cascade slopes up to the crest in the NOCRW simulation was also 10%–20% greater than that in the NOCR simulation and 20%–30% greater than in the CTL (Fig. 20). The simulated integrated water vapor at the UW site (just west of the Cascades) was 5% and 15% greater in the NOCR and NOCRW runs, respectively, than for the CTL run (not shown). The NOCRW run had more instability at the UW location (not shown), which favored increased convective activity over the adjacent slopes of the Cascades. The NOCRW simulation shows a gradual increase of the precipitation from the coast toward the Cascade crest, suggesting that this barrier influences the precipitation well upstream (∼170 km westward) of the crest.

6. Conclusions

This paper describes the coevolving kinematic and precipitation fields over the Oregon Coast Range and Cascades during the passage of a notably cold, unstable trough on 4–5 December 2001 during the second Improvement of Microphysical Parameterization through Observational Verification Experiment field project. This event featured a landfalling baroclinic wave, but, unlike previously documented IMPROVE-2 events, this trough developed within a colder air mass and was accompanied by comparatively weak cross-barrier flow and thermodynamic stratification. The low-level flow response was complex, since there was sufficient stability below 700 m upstream of the Cascades to favor channeled southerly near-surface winds, while farther aloft the flow was increasingly directed up and over the Cascades. Unlike most landfalling baroclinic waves impacting the Pacific Northwest, the initial phase of this storm was marked by relatively weak warm advection at low-levels, so the precipitation did not begin in the more typical mode of a lowering stratiform shield described by Medina et al. (2007). Over the coast, the onshore flow at low levels was relatively unstable, and supported embedded convective cells that were preferentially enhanced as they made landfall and encountered the more modest coastal mountains. These cells propagated eastward and briefly weakened while passing over the Willamette Valley before reintensifying upon arrival over the windward (west facing) slopes of the Cascades. As the upper-level trough approached, precipitation increased in coverage and intensity, and greater precipitation spillover into the lee of the Cascades accompanied increasing cross-barrier flow and overall orographic cloud depth. What makes this case unique is that there was a superposition of a steady orographically forced zones of ascent combined with more transient cellular convective precipitation that initially formed and/or intensified near the elevated coastal terrain before advancing eastward and subsequently interacting with the Cascade barrier some ∼100–150 km downstream.

The Weather Research and Forecasting Model (version 2.2) was used to simulate the synoptic and mesoscale structures of this event down to horizontal 1.33-km grid spacing. The model was verified using numerous observational platforms including in situ aircraft measurements, S-Pol radar, wind profilers, radiosondes, and surface observations. The model realistically simulated the three-dimensional thermodynamic, kinematic, and precipitation structures although it underestimated the cross-barrier flow by a few meters per second as well as the depth and strength of an associated shear layer over the windward slopes. The WRF realistically predicted the ∼1 m s−1 vertical velocities associated with relatively weak gravity waves over the narrow (10–20 km) ridges located on the windward (western) face of the Cascades; however it overestimated cloud water production over these ridges. The 1.33-km grid run simulated precipitation accumulations to within 20% of those observed at many stations over the Cascades, which is in marked contrast to the characteristic tendency for overprediction noted in previous IMPROVE-2 studies of storms whose environments were characterized by stronger cross-barrier flow and more stable stratification.

The flow over the small-scale windward ridges combined with the embedded convection increased the precipitation locally over these windward ridges by nearly 50% as compared to adjacent valleys. Even so, a sensitivity run employing an artificially smoothed version of the Cascades illustrates than the narrow ridges had a relatively small (<5%) impact on the net precipitation given the lack of riming (because of the relatively weak vertical motions over the narrow ridges) and the relatively low freezing level. A sensitivity run eliminating elevated coastal terrain and associated land–sea contrasts in surface friction showed that the broader circulation tied to the larger mountain wave anchored over the Cascade crest enhanced the precipitation well (∼100 km) upstream of this more major barrier.

Model sensitivity experiments have further illuminated the importance of the Coast Range in triggering convective cells and associated local precipitation enhancement. In this case the net effect of the coastal range had two components: one was the lifting of the flow over the elevated terrain profile and another was the low-level convergence induced by the land–sea frictional contrasts across the coastline. These two sources of forcing contributed to a significant (20%–30%) decrease of the precipitation over the Cascades, which lay downwind of this coastal zone. Sensitivity experiments show that the coastal terrain influences the stratiform versus convective nature and intensity of the precipitation farther downstream by 1) removing moisture from the landfalling low-level airstream and 2) lifting the airflow over the coastal terrain and hence triggering convection which subsequently reduces the of amount convective instability available downstream.

A subsequent paper will highlight the observed and simulated cloud microphysics for the 4–5 December event. We are also currently evaluating the WRF model more extensively using two cool seasons of Weather Surveillance Radar-1988 Doppler (WSR-88D) and surface data over northern Oregon in order to put these IMPROVE results in proper perspective relative to the full range of observed storm types.

Acknowledgments

This research was supported by the National Science Foundation (ATM-0450444). We thank Neal Johnson for his help in making the observed mesoscale surface map. Amy Haase skillfully edited the airborne Doppler radar data. We appreciate the constructive comments by the three reviewers to help improve the manuscript. Use of the WRF was made possible by the Microscale and Mesoscale Meteorological (MMM) Division of the National Center for Atmospheric Research (NCAR), which is supported by the National Science Foundation. The third author was supported by NSF ATM-0505739.

REFERENCES

  • Asencio, N., , J. Stein, , M. Chong, , and F. Gheusi, 2003: Analysis and simulation of local and regional conditions for the rainfall over the Lago Maggiore target area during MAP IOP2b. Quart. J. Roy. Meteor. Soc., 129 , 565586.

    • Search Google Scholar
    • Export Citation
  • Bougeault, P., and Coauthors, 2001: The MAP Special Observing Period. Bull. Amer. Meteor. Soc., 82 , 433462.

  • Bousquet, O., , and B. F. Smull, 2003a: Airflow and precipitation fields within deep Alpine valleys observed by airborne Doppler radar. J. Appl. Meteor., 42 , 14971513.

    • Search Google Scholar
    • Export Citation
  • Bousquet, O., , and B. F. Smull, 2003b: Observations and impacts of upstream blocking during a widespread orographic precipitation event. Quart. J. Roy. Meteor. Soc., 129 , 391409.

    • Search Google Scholar
    • Export Citation
  • Braun, S. A., 2006: High-resolution simulation of Hurricane Bonnie (1998). Part II: Water budget. J. Atmos. Sci., 63 , 4364.

  • Braun, S. A., , R. Rotunno, , and J. B. Klemp, 1999: Effects of coastal orography on landfalling cold fronts. Part II: Effects of surface friction. J. Atmos. Sci., 56 , 33663384.

    • Search Google Scholar
    • Export Citation
  • Browning, K. A., , F. F. Hill, , and C. W. Pardoe, 1974: Structure and mechanism of precipitation and the effect of orography in a wintertime warm sector. Quart. J. Roy. Meteor. Soc., 100 , 309330.

    • Search Google Scholar
    • Export Citation
  • Bruintjes, R. T., , T. L. Clark, , and W. D. Hall, 1994: Interactions between topographic airflow and cloud/precipitation development during the passage of a winter storm in Arizona. J. Atmos. Sci., 51 , 4867.

    • Search Google Scholar
    • Export Citation
  • Colle, B. A., 2004: Sensitivity of orographic precipitation to changing ambient conditions and terrain geometries: An idealized modeling perspective. J. Atmos. Sci., 61 , 588606.

    • Search Google Scholar
    • Export Citation
  • Colle, B. A., 2008: Two-dimensional idealized simulations of the impact of multiple windward ridges on orographic precipitation. J. Atmos. Sci., 65 , 509523.

    • Search Google Scholar
    • Export Citation
  • Colle, B. A., , and Y. Zeng, 2004: Bulk microphysical sensitivities within the MM5 for orographic precipitation. Part II: Impact of barrier width and freezing level. Mon. Wea. Rev., 132 , 28022815.

    • Search Google Scholar
    • Export Citation
  • Colle, B. A., , and S. E. Yuter, 2007: The impact of coastal boundaries and small hills on the precipitation distribution across southern Connecticut and Long Island, New York. Mon. Wea Rev., 135 , 933954.

    • Search Google Scholar
    • Export Citation
  • Colle, B. A., , C. F. Mass, , and K. W. Westrick, 2000: MM5 precipitation verification over the Pacific Northwest during the 1997–99 cool seasons. Wea. Forecasting, 15 , 730744.

    • Search Google Scholar
    • Export Citation
  • Colle, B. A., , J. B. Wolfe, , W. J. Steenburgh, , D. E. Kingsmill, , J. A. Cox, , and J. C. Shafer, 2005: High resolution simulations and microphysical validation of an orographic precipitation event over the Wasatch Mountains during IPEX IOP3. Mon. Wea. Rev., 133 , 29472971.

    • Search Google Scholar
    • Export Citation
  • Cox, J. A. W., , W. J. Steenburgh, , D. E. Kingsmill, , J. C. Shafer, , B. A. Colle, , O. Bousquet, , B. F. Smull, , and H. Cai, 2005: The kinematic structure of a Wasatch Mountain winter storm during IPEX IOP3. Mon. Wea. Rev., 133 , 521542.

    • Search Google Scholar
    • Export Citation
  • Durran, D., , and J. Klemp, 1982: On the effects of moisture on the Brunt–Väisälä frequency. J. Atmos. Sci., 39 , 21522158.

  • Garvert, M. F., , B. A. Colle, , and C. F. Mass, 2005a: 13–14 December 2001 IMPROVE-2 event. Part I: Synoptic and mesoscale evolution and comparison with a mesoscale model simulation. J. Atmos. Sci., 62 , 34743492.

    • Search Google Scholar
    • Export Citation
  • Garvert, M. F., , C. Woods, , B. A. Colle, , C. F. Mass, , and P. Hobbs, 2005b: 13–14 December 2001 IMPROVE-2 event. Part II: Evaluation of cloud and precipitation structures in the MM5 model simulation. J. Atmos. Sci., 62 , 35203534.

    • Search Google Scholar
    • Export Citation
  • Garvert, M. F., , B. F. Smull, , and C. F. Mass, 2007: Multiscale mountain waves influencing a major orographic precipitation event. J. Atmos. Sci., 64 , 711737.

    • Search Google Scholar
    • Export Citation
  • Georgis, J-F., , F. Roux, , M. Chong, , and S. Pradier, 2003: Triple-Doppler radar analysis of the heavy rain event observed in the Lago Maggiore region during MAP IOP2b. Quart. J. Roy. Meteor. Soc., 129 , 495522.

    • Search Google Scholar
    • Export Citation
  • Hill, F. F., , K. A. Browning, , and M. J. Bader, 1981: Radar and raingauge observations of orographic rain over South Wales. Quart. J. Roy. Meteor. Soc., 107 , 643670.

    • Search Google Scholar
    • Export Citation
  • Houze, R., , and S. Medina, 2005: Turbulence as a mechanism for orographic precipitation enhancement. J. Atmos. Sci., 62 , 35993623.

  • Janjić, Z. I., 1994: The step-mountain eta coordinate model: Further developments of the convection, viscous sublayer, and turbulence closure schemes. Mon. Wea. Rev., 122 , 927945.

    • Search Google Scholar
    • Export Citation
  • Kain, J. S., 2004: The Kain–Fritsch convective parameterization: An update. J. Appl. Meteor., 43 , 170181.

  • Kirshbaum, D. J., , and D. R. Durran, 2005: Atmospheric factors governing banded orographic convection. J. Atmos. Sci., 62 , 37583774.

  • Koch, S. E., , B. S. Ferrier, , M. T. Stoelinga, , E. J. Szoke, , S. J. Weiss, , and J. S. Kain, 2005: The use of simulated radar reflectivity fields in the diagnosis of mesoscale phenomena from high-resolution WRF model forecasts. Preprints, 32nd Conf. on Radar Meteorology, Albuquerque, NM, Amer. Meteor. Soc., J4J.7. [Available online at http://ams.confex.com/ams/pdfpapers/97032.pdf.].

  • Marwitz, J. D., 1987a: Deep orographic storms over the Sierra Nevada. Part I: Thermodynamic and kinematic structure. J. Atmos. Sci., 44 , 159173.

    • Search Google Scholar
    • Export Citation
  • Marwitz, J. D., 1987b: Deep orographic storms over the Sierra Nevada. Part II: The precipitation processes. J. Atmos. Sci., 44 , 174185.

    • Search Google Scholar
    • Export Citation
  • Medina, S., , and R. Houze, 2003: Air motions and precipitation growth in alpine storms. Quart. J. Roy. Meteor. Soc., 129 , 345371.

  • Medina, S., , B. Smull, , R. Houze, , and M. Steiner, 2005: Cross-barrier flow during orographic precipitation events: Results from MAP and IMPROVE. J. Atmos. Sci., 62 , 35803598.

    • Search Google Scholar
    • Export Citation
  • Medina, S., , E. Sukovich, , and R. A. Houze Jr., 2007: Vertical structures of precipitation in cyclones crossing the Oregon Cascades. Mon. Wea. Rev., 135 , 35653586.

    • Search Google Scholar
    • Export Citation
  • Rotunno, R., , and R. Ferretti, 2001: Mechanisms of intense alpine rainfall. J. Atmos. Sci., 58 , 17321749.

  • Rotunno, R., , and R. A. Houze Jr., 2007: Lessons on orographic precipitation from the Mesoscale Alpine Programme. Quart. J. Roy. Meteor. Soc., 133 , 811830.

    • Search Google Scholar
    • Export Citation
  • Schultz, D. M., and Coauthors, 2002: Understanding Utah winter storms: The Intermountain Precipitation Experiment. Bull. Amer. Meteor. Soc., 83 , 189210.

    • Search Google Scholar
    • Export Citation
  • Skamarock, W. C., 2006: Positive-definite and monotonic limiters for unrestricted-time-step transport schemes. Mon. Wea. Rev., 134 , 22412250.

    • Search Google Scholar
    • Export Citation
  • Skamarock, W. C., , J. B. Klemp, , J. Dudhia, , D. O. Gill, , D. M. Barker, , W. Wang, , and J. G. Powers, 2005: A description of the Advanced Research WRF, version 2. NCAR Tech. Note NCAR/TN-468+STR, 88 pp. [Available from UCAR Communications, P.O. Box 3000, Boulder, CO 80307].

  • Smith, R. B., , and I. Barstad, 2004: A linear theory of orographic precipitation. J. Atmos. Sci., 61 , 13771391.

  • Smith, R. B., , Q. Jiang, , M. Fearon, , P. Tabary, , M. Dorninger, , and J. Doyle, 2003: Orographic precipitation and air mass transformation: An alpine example. Quart. J. Roy. Meteor. Soc., 129 , 433454.

    • Search Google Scholar
    • Export Citation
  • Stoelinga, M. T., and Coauthors, 2003: Improvement of microphysical parameterization through observational verification experiment. Bull. Amer. Meteor. Soc., 84 , 18071826.

    • Search Google Scholar
    • Export Citation
  • Thompson, G., , R. M. Rasmussen, , and K. Manning, 2004: Explicit forecasts of winter precipitation using an improved bulk microphysics scheme. Part I: Description and sensitivity analysis. Mon. Wea. Rev., 132 , 519542.

    • Search Google Scholar
    • Export Citation
  • White, A. B., , P. J. Neiman, , F. M. Ralph, , D. E. Kingsmill, , and P. O. G. Persson, 2003: Coastal orographic rainfall processes observed by radar during the California land-falling jets experiment. J. Hydrometeor., 4 , 264282.

    • Search Google Scholar
    • Export Citation
Fig. 1.
Fig. 1.

(a) WRF analysis of 500-mb geopotential height (solid contours every 60 m), temperature (dashed contours every 5°C), wind barbs (1 full barb = 10 kt) and 300-mb isotachs (shaded light, medium, and dark gray for 50, 60, and 70 m s−1, respectively) from the 36-km WRF at 1200 UTC 4 Dec 2001. Inset boxes denote nested 12-, 4-, and 1.33-km WRF domains. (b) Terrain (shaded, m) and NOAA P-3 flight legs on 4–5 Dec. Dashed lines AB and CD indicate the orientations of the S-Pol vertical cross sections shown in Figs. 9, 10 and 16.

Citation: Monthly Weather Review 136, 10; 10.1175/2008MWR2369.1

Fig. 2.
Fig. 2.

Geostationary Operational Environmental Satellite (GOES) infrared satellite images (K, shaded according to inset scale) at (a) 1200 UTC 4 Dec 2001 and (b) 0000 UTC 5 Dec 2001.

Citation: Monthly Weather Review 136, 10; 10.1175/2008MWR2369.1

Fig. 3.
Fig. 3.

(a) NCEP analysis of 500-mb geopotential height (solid every 60 m), temperature (dashed every 5°C), and winds (1 full barb = 10 kt) at 0000 UTC 5 Dec 2001. (b) As in (a) but for the 12-h WRF 36-km simulation and showing the wind speed at 300-mb (light, medium, and dark gray shading denotes speeds exceeding 50, 60, and 70 m s−1).

Citation: Monthly Weather Review 136, 10; 10.1175/2008MWR2369.1

Fig. 4.
Fig. 4.

(a) NCEP analysis of 850-mb geopotential height (solid every 30 m), temperature (dashed every 5°C), and winds (1 full barb = 10 kt) at 0000 UTC 5 Dec 2001. (b) As in (a) but for the 12-h WRF 36-km simulation. (c) As in (a) but at 1200 UTC 5 Dec. (d) As in (c) but for the 12-h WRF 36-km simulation initialized at 0000 UTC 5 Dec.

Citation: Monthly Weather Review 136, 10; 10.1175/2008MWR2369.1

Fig. 5.
Fig. 5.

(a) Surface observations over western and central Oregon at 0000 UTC 5 Dec (using std surface model, full barb = 10 kt) (b) As in (a) but for the corresponding 4-km WRF simulation showing sea level pressure (solid every 1 mb), surface temperature (dashed every 2°C), and terrain height (shaded every 500 m).

Citation: Monthly Weather Review 136, 10; 10.1175/2008MWR2369.1

Fig. 6.
Fig. 6.

(a) Observed time–height series of equivalent potential temperature (solid) every 1 K using the UW and SLE soundings at (day/time) 04/1200, 04/1630, 05/0000, 05/0130, 05/0330, 05/0600, 05/1200 UTC, as well as winds at 1-h intervals from the IB profiler. (b) As in (a), but for the corresponding 1.33-km WRF simulation output of winds at location IB and equivalent potential temperatures at UW.

Citation: Monthly Weather Review 136, 10; 10.1175/2008MWR2369.1

Fig. 7.
Fig. 7.

Vertical profiles of (a) cross-barrier (west to east) flow (m s−1), (b) potential temperature (short dashed, K) and equivalent potential temperature (solid, K), and (c) the square of moist static stability (N2m) at the UW sounding site at 2100 UTC 4 Dec. Observed (1.33-km WRF-simulated) quantities at the UW site are shown in solid black (solid gray). The 1.33-km WRF moist static stability at point A (Fig. 8b) is shown by the dashed gray line in (c). (d)–(f) As in (a)–(c) but for 0330 UTC 5 Dec.

Citation: Monthly Weather Review 136, 10; 10.1175/2008MWR2369.1

Fig. 8.
Fig. 8.

Observed S-Pol reflectivities (shaded, 1.5° scan) and terrain (solid, 1.0 and 2.0 km MSL contours) at (a) 2000 UTC 4 Dec (c) 0000 UTC 5 Dec, (e) 0400 UTC 5 Dec, and (g) 0800 UTC 5 Dec. The range ring spacing is 50 km. WRF-derived reflectivity (700 mb) from the 1.33-km simulation at (b) 2000 UTC 4 Dec and (d) 0000 UTC, (f) 0400 UTC, and (h) 0800 UTC 5 Dec using gray-shaded scale (every 6 dBZ). The WRF wind vectors at 700 mb are also shown as well as the model terrain (solid, 1.0 and 2.0 m contours). Line AB in (b) is the location for the RHI scans in Figs. 9, 10.

Citation: Monthly Weather Review 136, 10; 10.1175/2008MWR2369.1

Fig. 9.
Fig. 9.

Observed S-Pol reflectivities for cross section AB (shown in Fig. 8b) at 2000 UTC 4 Dec using gray-shaded scale (every 6 dBZ) for (a) 2000 UTC 4 Dec, (c) 0000 UTC 5 Dec, (e) 0400 UTC 5 Dec, and (g) 0800 UTC 5 Dec. The WRF reflectivities, potential temperature (solid every 3 K), and circulation vectors [m s−1 using the inset scale in (b)] from the 1.33-km simulation are shown for (b) 2000 UTC, (d) 0000 UTC, (f) 0400 UTC, and (h) 0800 UTC 5 Dec.

Citation: Monthly Weather Review 136, 10; 10.1175/2008MWR2369.1

Fig. 10.
Fig. 10.

Observed S-Pol reflectivities for cross section AB (shown in Fig. 8b) at (a) 0013 UTC, (b) 0035 UTC, (c) 0056 UTC, (d) 0118 UTC, (e) 0139 UTC, and (f) 0200 UTC 5 Dec using gray-shaded scale (every 6 dBZ). The arrows and numbers indicate the cell locations as they propagate from west to east across the Coast Range and Cascades.

Citation: Monthly Weather Review 136, 10; 10.1175/2008MWR2369.1

Fig. 11.
Fig. 11.

(a) Hovmöller-type plot showing the maximum observed S-Pol radar reflectivity (color shaded every 2.5 dBZ) above 2.0 km MSL from 1900 UTC 4 Dec to 1200 UTC 5 Dec along AB (see Fig. 8b for location). (b) As in (a) but for the 1.33-km WRF.

Citation: Monthly Weather Review 136, 10; 10.1175/2008MWR2369.1

Fig. 12.
Fig. 12.

(a) Observed NOAA P-3 reflectivity (shaded every 6 dBZ) at 1.5 km MSL for legs 2–4 between 2300 UTC 4 Dec and 0100 UTC 5 Dec. There are no data for much of the lee of the Cascades. (b) As in (a) but for the 1.33-km WRF at 0000 UTC 5 Dec as well as the wind vectors at 1.5 km MSL. The boxed region in (a) shows the region included in deriving average cross-barrier wind speeds shown in Fig. 13. The terrain is also shown (0.5 km contours).

Citation: Monthly Weather Review 136, 10; 10.1175/2008MWR2369.1

Fig. 13.
Fig. 13.

(a) Observed wind speeds (shaded every 3 m s−1) and barrier-parallel winds (υ component, dashed every 3 m s−1) from NOAA P-3 Doppler radar averaged over the west–east box across the Cascades for legs 2–4 (boxed region shown in Fig. 12a) between 2351 UTC 4 Dec and 0055 UTC 5 Dec. (b) As in (a) but for the 1.33-km WRF at 0030 UTC 5 Dec. (c) As in (a) but for leg 6 at 0144–0200 UTC 5 Dec. (d) As in (c) but for the 1.33-km WRF at 0200 UTC 5 Dec.

Citation: Monthly Weather Review 136, 10; 10.1175/2008MWR2369.1

Fig. 14.
Fig. 14.

(a) Simulated precipitation (shaded every 5 mm) from the 1.33-km WRF between 2000 UTC 4 Dec to 0800 UTC 5 Dec and terrain (0.5 km contours). (b) Percent of observed precipitation simulated by the 1.33-km WRF at the precipitation gauge sites between 2000 UTC 4 Dec and 0800 UTC 5 Dec. The percent values are color coded using the inset scale in (b), while background shading indicates terrain elevation (m, key at lower left).

Citation: Monthly Weather Review 136, 10; 10.1175/2008MWR2369.1

Fig. 15.
Fig. 15.

Observed vertically pointing radar reflectivity (shaded every 6 dBZ) from the S-Prof (location MB on Fig. 1b) from 1200 UTC 4 Dec to 1200 UTC 5 Dec.

Citation: Monthly Weather Review 136, 10; 10.1175/2008MWR2369.1

Fig. 16.
Fig. 16.

(a) Observed S-Pol radar reflectivity for cross section AB (shown in Fig. 8b) averaged between 0200 UTC and 0600 UTC 5 Dec using gray-shaded scale (every 6 dBZ). (b) S-Pol reflectivity for cross section CD (shown in Fig. 1b) averaged between 2100 UTC 13 Dec and 0100 UTC 14 Dec using gray-shaded scale (every 6 dBZ).

Citation: Monthly Weather Review 136, 10; 10.1175/2008MWR2369.1

Fig. 17.
Fig. 17.

Time series from NOAA P-3, 4-km WRF, and 1.33-km WRF along leg 2 from south (left) to north (right) at 2450 m MSL showing (a) wind speed (m s−1), (b) cloud water (g m−3), vertical velocity (cm s−1), and terrain (m) between 2351 UTC 4 Dec and 0008 UTC 5 Dec.

Citation: Monthly Weather Review 136, 10; 10.1175/2008MWR2369.1

Fig. 18.
Fig. 18.

(a) Observed NOAA P-3 cross section of radar reflectivity (every 2.5 dBZ) and dual-Doppler vertical motions (white contour every 0.3 m s−1) along leg 2 from south (left) to north (right) at 2400 m MSL between 2351 UTC 4 Dec and 0008 UTC 5 Dec. (b) As in (a) but for the 1.33-km WRF simulation output showing circulation vectors in the cross section, cloud water (shaded in g kg−1), snow (solid every 0.1 g kg−1), and graupel (dashed every 0.1 g kg−1). The terrain height within the plane of the section is shown at the bottom of each panel.

Citation: Monthly Weather Review 136, 10; 10.1175/2008MWR2369.1

Fig. 19.
Fig. 19.

Simulated precipitation difference (mm) for the 1.33-km WRF run of (a) SMTH–CTL, (b) NOCR – CTL, and (c) NOCRW – CTL between 2000 UTC 4 Dec and 0800 UTC 5 Dec. The gray shading (every 500 m) in (a) denotes the SMTH terrain, while the shading in (b) and (c) represents the NOCR and NOCRW terrain, respectively. The dashed box in (a) represents the region in which the precipitation was averaged in latitude for Fig. 20.

Citation: Monthly Weather Review 136, 10; 10.1175/2008MWR2369.1

Fig. 20.
Fig. 20.

Simulated precipitation (mm) averaged in latitude over the boxed area in Fig. 19a for the CTL, SMTH, NOCR, and NOCRW runs for the (a) 2000 UTC 4 Dec–0200 UTC 5 Dec and (b) 0200–0600 UTC 5 Dec periods. The terrain profile is indicated at the bottom of the panels. The tables at the top of the panels indicate averages over the coastal range, windward slope, and crest–lee.

Citation: Monthly Weather Review 136, 10; 10.1175/2008MWR2369.1

Fig. 21.
Fig. 21.

Hovmöller plot showing the vertically integrated water and ice (color shaded, mm) for the 1.33-km WRF from 1900 UTC 4 Dec to 0600 UTC 5 Dec along AB (see Fig. 8b for location) for (a) CTL, (b) SMTH, (c) NOCR, and (d) NOCRW experiments. The terrain profile for the CTL run is shown in the lower panel.

Citation: Monthly Weather Review 136, 10; 10.1175/2008MWR2369.1

1

Model-simulated reflectivities were calculated using empirical relations based on the model cloud and precipitation mixing ratios as in Koch et al. (2005), with modifications to the variable slope intercept for snow based on Thompson et al. (2004).

2

It is possible that the observed values of maximum reflectivity are underestimated at ranges >75 km from the radar (where the base of the echo intersects the 2.0 km MSL). However this scenario is unlikely given the large degree of embedded convection that was observed in this case. The cells over the coast and farther upstream have high echo tops that extend well above 3–4 km MSL, that is, to levels where the reflectivity is well sampled even at far ranges (Figs. 9, 10).

3

The pulsing cell intensity in the simulated output is an artifact of constructing the plot using 10-min output to match the frequency of the S-Pol data (available only every 11 min), which causes artificial gaps in the reflectivity intensity versus time. Since the simulation generated more cells than observed over the coastal range (Figs. 8c,d), the gaps between the WRF-simulated cells are more pronounced than in the S-Pol observations.

Save