• Adams, D. K., , and A. C. Comrie, 1997: The North American monsoon. Bull. Amer. Meteor. Soc., 78 , 21972213.

  • Anderson, B. T., , and H. Kanamaru, 2005: The diurnal cycle of the summertime atmospheric hydrologic cycle over the southwestern United States. J. Hydrometeor., 6 , 219228.

    • Search Google Scholar
    • Export Citation
  • Anderson, B. T., , J. O. Roads, , and S-C. Chen, 2000: Large-scale forcing of summertime monsoon surges over the Gulf of California and the southwestern United States. J. Geophys. Res., 105 , 2445524467.

    • Search Google Scholar
    • Export Citation
  • Anderson, B. T., , J. O. Roads, , S-C. Chen, , and H-M. H. Juang, 2001: Model dynamics of summertime low-level jets over northwestern Mexico. J. Geophys. Res., 106 , 34013413.

    • Search Google Scholar
    • Export Citation
  • Barlow, M., , S. Nigam, , and E. H. Berbery, 1998: Evolution of the North American monsoon system. J. Climate, 11 , 22382257.

  • Berbery, E. H., 2001: Mesoscale moisture analysis of the North American monsoon. J. Climate, 14 , 121137.

  • Berbery, E. H., , and M. S. Fox-Rabinovitz, 2003: Multiscale diagnosis of the North American monsoon system using a variable-resolution GCM. J. Climate, 16 , 19291947.

    • Search Google Scholar
    • Export Citation
  • Bordoni, S., , P. E. Ciesielski, , R. H. Johnson, , B. D. McNoldy, , and B. Stevens, 2004: The low-level circulation of the North American Monsoon as revealed by QuikSCAT. Geophys. Res. Lett., 31 .L10109, doi:10.1029/2004GL020009.

    • Search Google Scholar
    • Export Citation
  • Douglas, M. W., 1995: The summertime low-level jet over the Gulf of California. Mon. Wea. Rev., 123 , 23342348.

  • Douglas, M. W., , and J. C. Leal, 2003: Summertime surges over the Gulf of California: Aspects of their climatology, mean structure, and evolution from radiosonde, NCEP reanalysis, and rainfall data. Wea. Forecasting, 18 , 5574.

    • Search Google Scholar
    • Export Citation
  • Douglas, M. W., , R. A. Maddox, , and K. Howard, 1993: The Mexican monsoon. J. Climate, 6 , 16651677.

  • Douglas, M. W., , A. Valdez-Manzanilla, , and R. G. Cueto, 1998: Diurnal variation and horizontal extent of the low-level jet over the northern Gulf of California. Mon. Wea. Rev., 126 , 20172025.

    • Search Google Scholar
    • Export Citation
  • Farfán, L. M., , and J. A. Zehnder, 1994: Moving and stationary mesoscale convective systems over northwest Mexico during the Southwest Area Monsoon Project. Wea. Forecasting, 9 , 630639.

    • Search Google Scholar
    • Export Citation
  • Fawcett, P. J., , J. R. Stalker, , and D. S. Gutzler, 2002: Multistage moisture transport into the interior of northern Mexico during the North American summer monsoon. Geophys. Res. Lett., 29 .2094, doi:10.1029/2002GL015693.

    • Search Google Scholar
    • Export Citation
  • Fuller, R. D., , and D. J. Stensrud, 2000: The relationship between tropical easterly waves and surges over the Gulf of California during the North American monsoon. Mon. Wea. Rev., 128 , 29832989.

    • Search Google Scholar
    • Export Citation
  • Gochis, D. J., , J-C. Leal, , C. J. Watts, , W. J. Shuttleworth, , and J. Gartuza-Payan, 2003: Preliminary diagnostics from a new event-based precipitation monitoring system in support of the North American Monsoon Experiment. J. Hydrometeor., 4 , 974981.

    • Search Google Scholar
    • Export Citation
  • Gochis, D. J., , A. Jimenez, , C. J. Watts, , J. Garatuza-Payan, , and W. J. Shuttleworth, 2004: Analysis of 2002 and 2003 warm-season precipitation from the North American Monsoon Experiment event rain gauge network. Mon. Wea. Rev., 132 , 29382953.

    • Search Google Scholar
    • Export Citation
  • Hales Jr., J. E., 1972: Surges of maritime tropical air northward over the Gulf of California. Mon. Wea. Rev., 100 , 298306.

  • Hales Jr., J. E., 1974: Southwestern United States summer monsoon source—Gulf of Mexico or Pacific Ocean? J. Appl. Meteor., 13 , 331342.

    • Search Google Scholar
    • Export Citation
  • Higgins, R. W., , and W. Shi, 2005: Relationships between Gulf of California moisture surges and tropical cyclones in the eastern Pacific basin. J. Climate, 18 , 46014620.

    • Search Google Scholar
    • Export Citation
  • Higgins, R. W., , Y. Yao, , and X. L. Wang, 1997: Influence of the North American monsoon system on the U.S. summer precipitation regime. J. Climate, 10 , 26002621.

    • Search Google Scholar
    • Export Citation
  • Higgins, R. W., , W. Shi, , E. Yarosh, , and R. Joyce, 2000: Improved United States Precipitation Quality Control System and Analysis. NCEP/Climate Prediction Center Atlas 7, 40 pp. [Available online at http://www.cpc.ncep.noaa.gov/research_papers/ncep_cpc_atlas/7/index.html.].

    • Search Google Scholar
    • Export Citation
  • Higgins, R. W., , W. Shi, , and C. Hain, 2004: Relationships between Gulf of California moisture surges and precipitation in the southwestern United States. J. Climate, 17 , 29832995.

    • Search Google Scholar
    • Export Citation
  • Higgins, R. W., and Coauthors, 2006: The NAME 2004 field campaign and modeling strategy. Bull. Amer. Meteor. Soc., 87 , 7994.

  • Howard, K. W., , and R. A. Maddox, 1988: Mexican mesoscale convective systems—A satellite perspective. Preprints, Third Int. American and Mexican Congress of Meteorology, Mexico City, Mexico, Mexican Meteorological Organization, 404–408.

  • Janowiak, J. E., , P. Xie, , R. Joyce, , M. Chen, , and Y. Yarosh, 2004: Validation of daily satellite precipitation estimates over the U.S. Proc. 29th Annual Climate Diagnostics and Prediction Workshop, Madison, WI, NOAA. [Available online at http://www.cpc.noaa.gov/products/outreach/proceedings/cdw29_proceedings/CDW29.proceedings.shtml.].

    • Search Google Scholar
    • Export Citation
  • Janowiak, J. E., , V. J. Dagostaro, , V. E. Kousky, , and R. J. Joyce, 2007: An examination of precipitation in observations and model forecasts during NAME with emphasis on the diurnal cycle. J. Climate, 20 , 16801692.

    • Search Google Scholar
    • Export Citation
  • Johnson, R. H., , P. E. Ciesielski, , B. D. McNoldy, , P. J. Rogers, , and R. K. Taft, 2007: Multiscale variability of the flow during the North American Monsoon Experiment. J. Climate, 20 , 16281648.

    • Search Google Scholar
    • Export Citation
  • Joyce, R. J., , J. E. Janowiak, , P. A. Arkin, , and P. Xie, 2004: CMORPH: A method that produces global precipitation estimates from passive microwave and infrared data at high spatial and temporal resolution. J. Hydrometeor., 5 , 487503.

    • Search Google Scholar
    • Export Citation
  • Kistler, R., and Coauthors, 2001: The NCEP–NCAR 50-Year Reanalysis: Monthly means CD-ROM and documentation. Bull. Amer. Meteor. Soc., 82 , 247267.

    • Search Google Scholar
    • Export Citation
  • Lang, T. J., , D. A. Ahijevych, , S. W. Nesbitt, , R. E. Carbone, , S. A. Rutledge, , and R. Cifelli, 2007: RADAR-observed characteristics of precipitating systems during NAME 2004. J. Climate, 20 , 17131733.

    • Search Google Scholar
    • Export Citation
  • McCollum, J. R., , W. F. Krajewski, , R. R. Ferraro, , and M. B. Ba, 2002: Evaluation of biases of satellite rainfall estimation algorithms over the continental United States. J. Appl. Meteor., 41 , 10651080.

    • Search Google Scholar
    • Export Citation
  • Mesinger, F., and Coauthors, 2006: North American Regional Reanalysis. Bull. Amer. Meteor. Soc., 87 , 343360.

  • Mo, K., , M. Chelliah, , M. L. Carrera, , R. W. Higgins, , and W. Ebisuzaki, 2005: Atmospheric moisture transport over the United States and Mexico as evaluated in the NCEP regional reanalysis. J. Hydrometeor., 6 , 710728.

    • Search Google Scholar
    • Export Citation
  • Rosenfeld, D., , and Y. Mintz, 1988: Evaporation of rain falling from convective cloud as derived from radar measurements. J. Appl. Meteor., 27 , 209215.

    • Search Google Scholar
    • Export Citation
  • Shafran, P., , J. Woollen, , W. Ebisuzaki, , W. Shi, , Y. Fan, , R. Grumbine, , and M. Fennessy, 2004: Observational data used for assimilation in the NCEP North American regional reanalysis. Preprints, 14th Conf. on Applied Climatology, Seattle, WA, Amer. Meteor. Soc., 1.4.

  • Smith, W. P., , and R. L. Gall, 1989: Tropical squall lines of the Arizona monsoon. Mon. Wea. Rev., 117 , 15531569.

  • Stensrud, D. J., , R. L. Gall, , S. L. Mullen, , and K. W. Howard, 1995: Model climatology of the Mexican monsoon. J. Climate, 8 , 17751794.

    • Search Google Scholar
    • Export Citation
  • Stensrud, D. J., , R. L. Gall, , and M. K. Nordquist, 1997: Surges over the Gulf of California during the Mexican monsoon. Mon. Wea. Rev., 125 , 417437.

    • Search Google Scholar
    • Export Citation
  • Yarosh, Y., , P. Xie, , M. Chen, , R. Joyce, , J. E. Janowiak, , and P. A. Arkin, 2005: Diurnal cycle in the North American monsoon. Bull. Amer. Meteor. Soc., 86 , 2628.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    Daily average precipitation (mm day−1) over the North American monsoon region from July and August 2004: (a) CMORPH satellite estimates, (b) CPC U.S.–Mexico daily rain gauge analysis, (c) RMORPH satellite estimates, (d) North American Regional Reanalysis, and (e) Eta Model forecasts. The core monsoon region is clearly visible along the western slopes of the Sierra Madre Occidental between approximately 20° and 30°N.

  • View in gallery

    Hovmoeller diagrams of the diurnal cycle of precipitation in the core monsoon region, averaged into 1° latitude bands. This figure displays every other latitude. The diurnal cycle is shown from local noon to local noon to show the complete cycle; the contour interval is 0.1 mm h−1; (a) RMORPH, (b) NARR, and (c) Eta Model forecasts.

  • View in gallery

    Speed of westward shift of precipitation over the core monsoon region, as obtained from the slope of the Hovmoeller diagrams of precipitation from RMORPH, NARR, and the Eta Model.

  • View in gallery

    Hovmoeller diagrams of the diurnal cycle of (left) CAPE and (right) the vertically integrated moisture flux convergence from NARR. Precipitation contours have been overlain. The CAPE contour interval is 100 J kg−1, MFC contour interval is 0.2 kg m−2, and the precipitation contour interval is 0.1 mm h−1. To highlight the diurnal variation, the daily average of CAPE, which ranges from about 3000 J kg−1 over the Gulf of California to about 200 J kg−1 over the coastal plain and western SMO, has been removed. Both CAPE and the moisture flux convergence have been smoothed by the application of a 9-point smoothing function.

  • View in gallery

    Land-only area average of several elements in the core monsoon region from (a) NARR and (b) the Eta Model for the 24-h period beginning at local midnight. The upper panels depict precipitation (solid line, mm h−1), moisture flux convergence (long dash, mm day−1), and evaporation (dotted line, mm h−1). The middle panels are CAPE (J kg−1), and the bottom panels are CIN (J kg−1).

  • View in gallery

    Same as in Fig. 5 but for the southwestern United States.

  • View in gallery

    Percentage of precipitation during surge events for (left) Eta Model forecasts and (right) RMORPH satellite estimates. Shaded areas represent greater than 50% of total precipitation occurring during surge events; white areas are less than 50%. The total amount of precipitation that fell during surges was compared to the total precipitation of the entire 2-month period to produce this figure.

  • View in gallery

    (a) Meridional moisture flux at 30°N, 114°W, the northern Gulf of California; (b) area-averaged precipitation for the southwestern United States; and (c) the core North American monsoon region (bottom) for 10 Jul–31 Aug 2005. Eta Model forecasts (solid) and NARR satellite estimates (dashed); surge times are indicated by shaded regions.

  • View in gallery

    As in Fig. 6 but for the southwestern United States, divided into surge and nonsurge times. (a), (b) The 4-yr climatology from NARR of this region divided into surge and nonsurge. (c), (d) The 2004 NAME field campaign period obtained from NARR. (e), (f) The 2004 NAME field campaign period, as represented in the Eta Model. The upper panels depict precipitation (solid line, mm h−1), moisture flux convergence (long dashed line, mm day−1), and evaporation (dotted line, mm h−1). The middle panels are CAPE (J kg−1), and the bottom panels are CIN (J kg−1).

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 12 12 2
PDF Downloads 5 5 2

The Diurnal Cycle of Precipitation over the North American Monsoon Region during the NAME 2004 Field Campaign

View More View Less
  • 1 Department of Atmospheric and Oceanic Science/ESSIC, University of Maryland, College Park, College Park, Maryland
© Get Permissions
Full access

Abstract

The structure of the diurnal cycle of warm-season precipitation and its associated fields during the North American monsoon are examined for the core monsoon region and for the southwestern United States, using a diverse set of observations, analyses, and forecasts from the North American Monsoon Experiment field campaign of 2004. Included are rain gauge and satellite estimates of precipitation, Eta Model forecasts, and the North American Regional Reanalysis (NARR). Daily rain rates are of about the same magnitude in all datasets with the exception of the Climate Prediction Center (CPC) Morphing (CMORPH) technique, which exhibits markedly higher precipitation values.

The diurnal cycle of precipitation within the core region occurs earlier in the day at higher topographic elevations, evolving with a westward shift of the maximum. This shift appears in the observations, reanalysis, and, while less pronounced, in the model forecasts. Examination of some of the fields associated with this cycle, including convective available potential energy (CAPE), convective inhibition (CIN), and moisture flux convergence (MFC), reveals that the westward shift appears in all of them, but more prominently in the latter.

In general, warm-season precipitation in southern Arizona and parts of New Mexico shows a strong effect due to northward moisture surges from the Gulf of California. A reported positive bias in the NARR northward winds over the Gulf of California limits their use with confidence for studies of the moist surges along the Gulf; thus, the analysis is complemented with operational analysis and the Eta Model short-term simulations. The nonsurge diurnal cycle of precipitation lags the CAPE maximum by 6 h and is simultaneous with a minimum of CIN, while the moisture flux remains divergent throughout the day. During surges, CAPE and CIN have modifications only to the amplitude of their cycles, but the moisture flux becomes strongly convergent about 6 h before the precipitation maximum, suggesting a stronger role in the development of precipitation.

Corresponding author address: Ernesto Hugo Berbery, Department of Atmospheric and Oceanic Science/ESSIC, University of Maryland, College Park, 3427 Computer and Space Sciences Bldg., College Park, MD 20742-2425. Email: berbery@atmos.umd.edu

Abstract

The structure of the diurnal cycle of warm-season precipitation and its associated fields during the North American monsoon are examined for the core monsoon region and for the southwestern United States, using a diverse set of observations, analyses, and forecasts from the North American Monsoon Experiment field campaign of 2004. Included are rain gauge and satellite estimates of precipitation, Eta Model forecasts, and the North American Regional Reanalysis (NARR). Daily rain rates are of about the same magnitude in all datasets with the exception of the Climate Prediction Center (CPC) Morphing (CMORPH) technique, which exhibits markedly higher precipitation values.

The diurnal cycle of precipitation within the core region occurs earlier in the day at higher topographic elevations, evolving with a westward shift of the maximum. This shift appears in the observations, reanalysis, and, while less pronounced, in the model forecasts. Examination of some of the fields associated with this cycle, including convective available potential energy (CAPE), convective inhibition (CIN), and moisture flux convergence (MFC), reveals that the westward shift appears in all of them, but more prominently in the latter.

In general, warm-season precipitation in southern Arizona and parts of New Mexico shows a strong effect due to northward moisture surges from the Gulf of California. A reported positive bias in the NARR northward winds over the Gulf of California limits their use with confidence for studies of the moist surges along the Gulf; thus, the analysis is complemented with operational analysis and the Eta Model short-term simulations. The nonsurge diurnal cycle of precipitation lags the CAPE maximum by 6 h and is simultaneous with a minimum of CIN, while the moisture flux remains divergent throughout the day. During surges, CAPE and CIN have modifications only to the amplitude of their cycles, but the moisture flux becomes strongly convergent about 6 h before the precipitation maximum, suggesting a stronger role in the development of precipitation.

Corresponding author address: Ernesto Hugo Berbery, Department of Atmospheric and Oceanic Science/ESSIC, University of Maryland, College Park, 3427 Computer and Space Sciences Bldg., College Park, MD 20742-2425. Email: berbery@atmos.umd.edu

1. Introduction

Rainfall during the North American monsoon accounts for a majority of the annual total in the southwestern United States and western Mexico. The precipitation produced during the monsoon accounts for as much as 80% of annual rainfall in the area of western Mexico along the Gulf of California and up to 60% in southern Arizona and New Mexico (Douglas et al. 1993). The monsoon is characterized by a reversal in pressure and wind patterns resulting from the seasonal heating of land in the Northern Hemisphere summer (Adams and Comrie 1997). Midtropospheric flow shifts from dry and westerly in the late spring to moist and easterly/southeasterly in July (Douglas et al. 1993); a distinct pattern of increased convection and precipitation in northwestern Mexico and the southwestern United States is the primary effect of the system.

The core monsoon region, lying along western Mexico between approximately 20° and 30°N, receives the most intense precipitation of the geographical area directly affected by the monsoon. This region exhibits a prominent diurnal cycle of precipitation due to the combined effects of the sea breeze, with upslope flows produced by diurnal heating and topographic forcing by the Sierra Madre Occidental (SMO) mountain range (Johnson et al. 2007). Low-level moisture evaporated from the Gulf of California is transported by this upslope flow to the core region (Stensrud et al. 1995), and the resulting moisture flux convergence over the western slopes of the SMO leads to heavy precipitation (Douglas et al. 1993; Stensrud et al. 1995; Berbery 2001). Mesoscale convective systems that form over the mountains of western Mexico and drift to the west throughout their life spans are also believed to contribute to the heavy precipitation during the monsoon season (Howard and Maddox 1988; Smith and Gall 1989; Farfán and Zehnder 1994). The general temporal structure of the diurnal cycle of precipitation in the core region of the monsoon is for precipitation to begin in the late afternoon, peak in the late evening to midnight, and reach a minimum in the morning (Gochis et al. 2003).

In the northern sector of the monsoon, moisture is transported into Arizona and New Mexico via a nocturnal/predawn low-level jet from the Gulf of California (Douglas 1995; Berbery 2001). Douglas (1995) identified the presence of this jet, characterized by southerly winds below about 900 hPa, on about 75% of the summer days in his analysis. The southerly wind velocity is strongest in the late evening to early morning hours (Douglas et al. 1998), when the boundary layer is most stable, and is likely produced by the east–west temperature gradient created by nighttime cooling of the slopes of the SMO (Anderson et al. 2001). This jet transports significant quantities of water into northwest Mexico and southern Arizona (Fawcett et al. 2002).

Another occasional yet important feature of the monsoon is the northward surge of relatively cool, moist air through the Gulf of California, often linking precipitation in the southwestern United States to tropical disturbances in the Pacific Ocean (Hales 1972, 1974). These surges are accompanied by decreases in temperature and increases in humidity and surface pressure in southern Arizona. Surges, as seen in the low-level wind field (Bordoni et al. 2004), usually originate in the southern Gulf of California and travel northward into the southwestern United States and are usually associated with the passage of an easterly wave (Fuller and Stensrud 2000); tropical cyclones approaching the Gulf can also create strong surges (Anderson et al. 2000; Higgins and Shi 2005). Moisture surges along the Gulf have a critical role for the development of precipitation in the southwestern United States. These surges have been found to accompany a majority of the precipitation in Arizona during July and August (e.g., Berbery and Fox-Rabinovitz 2003; Higgins et al. 2004) but have only a modest effect on precipitation along northwestern Mexico (Douglas and Leal 2003).

The North American Monsoon Experiment (NAME) field campaign of 2004 has provided a valuable opportunity to analyze the variability of the monsoon system at smaller scales than those allowed by previous datasets. NAME incorporates observational datasets (including satellite estimates and an extensive rain gauge network) and empirical and modeling studies and employs a tiered approach. Tier I covers the core monsoon area of northwestern Mexico, as well as southern Arizona and New Mexico, and is the focus area for this study. NAME has a stated goal of determining the sources and limits of predictability of warm-season precipitation over North America (Higgins et al. 2006). This can only be achieved if physical mechanisms are properly represented, even at the mesoscale, including the diurnal cycle.

The objective of this study is to investigate the main features of the North American monsoon diurnal cycle of precipitation, using remotely sensed precipitation estimates, rain gauge measurements, Eta Model forecasts, and regional reanalysis data. This article is structured as follows: Section 2 contains information about the model and data products used for the study, and Section 3 details the results for the main features of the diurnal cycle of precipitation, including the structure of the diurnal cycle, potential associated fields, and the effect of moisture surges, for both the core monsoon region and the southwestern United States. Section 4 presents a summary of the work and conclusions.

2. Empirical, model, and reanalysis products

The diurnal cycle of precipitation and associated fields during the NAME-enhanced observing period of July and August 2004 was analyzed using data from several sources. Precipitation data examined for this study include the “research-quality” (RMORPH) version of the Climate Prediction Center (CPC) Morphing method (CMORPH) satellite estimates and observations from the CPC U.S.–Mexico rain gauge network. Several fields, including precipitation and associated thermodynamic and dynamic fields, were obtained from the North American Regional Reanalysis (NARR) products and from model forecasts produced using National Centers for Environmental Prediction’s (NCEP) workstation version of the Eta Model, run at the University of Maryland.

a. Empirical precipitation data

The primary satellite-estimated precipitation data for this study comes from RMORPH, which produces high-resolution (0.25° latitude × 0.25° longitude) precipitation estimates using passive microwave satellite rainfall estimates propagated by motion vectors from geostationary satellite infrared cloud motion estimates (Joyce et al. 2004). For precipitation estimates over land, RMORPH uses the CPC daily rain gauge analysis (Higgins et al. 2000), disaggregated by CMORPH satellite estimates (Janowiak et al. 2007). Although there may be several reasons for the CMORPH overestimation of precipitation, previous studies have suggested that in semiarid regions this error can result from the fact that the estimate is drawn from the cloud-top characteristics, while large evaporation may occur before the rainfall reaches the surface, resulting in too much precipitation at the surface (see Rosenfeld and Mintz 1988; McCollum et al. 2002; Janowiak et al. 2004; W. Shi et al. 2005, personal communication).

Rain gauge observations from the CPC U.S.–Mexico daily precipitation analysis were also examined. CPC U.S.–Mexico is available as daily total precipitation values only. The CPC U.S.–Mexico daily analysis has 1° × 1° spatial resolution, with over 8000 rain gauges covering a domain that includes all of the United States and Mexico (Higgins et al. 2000). Mexican precipitation data for the CPC U.S.–Mexico analysis comes from about 900 rain gauges, monitored by the Mexican National Weather Service. For comparison purposes, data from the NAME Event Rain Gauge Network (NERN) were also obtained (but not shown). NERN consists of 87 tipping-bucket rain gauges, installed in 2002 and 2003 in roughly east–west transects along the Sierra Madre Occidental between approximately 23° and 30°N, for the purposes of NAME (Gochis et al. 2003, 2004). NERN is available in 3-h intervals.

b. Reanalysis and model products

The NARR is a long-term, dynamically consistent, high-resolution, high-frequency, atmospheric and land surface hydrology dataset for the North American domain (Mesinger et al. 2006). Precipitation in NARR is assimilated, and over the continental United States and Mexico is obtained by the disaggregation of a daily rain gauge analysis (see Shafran et al. 2004). Over the continental United States, the daily analysis is disaggregated to hourly using temporal weights derived from a 2.5° × 2.5° latitude–longitude analysis of hourly rain gauge data; over Mexico, the 24-h analysis is disaggregated using the T62 resolution NCEP–Department of Energy (DOE) Global Reanalysis 2 precipitation forecasts (Kistler et al. 2001; Mesinger et al. 2006). NARR has a 32-km, 45-layer resolution, and the 3-hourly output is used for the present study. This fine horizontal and vertical grid allows for the study of features, such as the Gulf of California, which the NCEP–National Center for Atmospheric Research (NCAR) or NCEP–DOE global reanalyses, with a resolution of about 2.5° latitude × 2.5° longitude, do not resolve. Originally completed for the period 1979–2002, NARR has been continued in near–real time, and the months of the NAME field campaign have recently become available. The seasonal evolution of the monsoon obtained from NARR has been found to be largely consistent with observations (Mo et al. 2005). It has also been found (Mo et al. 2005) that NARR systematically overestimates the water vapor transport by the Gulf of California low-level jet (GCLLJ). The too-strong meridional winds have largest bias on the northern Gulf of California. Nevertheless, NARR captures the diurnal cycle of the meridional wind quite well (Mo et al. 2005), with minimum wind speeds at about 1600 LT and maximum at about 0100 LT.

Model forecasts for the 2004 summer season were produced with the workstation version of NCEP’s Eta Model, which was run at 22-km grid spacing, with 45 vertical levels; the output is available at 3-h intervals. The model is initialized from the Global Forecasting System (GFS) analysis, and the domain covers the area between 15° and 45°N, 120° and 90°W—roughly the same domain as NAME tier II (Higgins et al. 2006). The 12–36-h forecasts are employed for this analysis, thus avoiding spinup effects in the first 12 h.

c. Daily rain rate from empirical, model, and reanalysis products

Figure 1 illustrates the average daily rainfall in the western Mexico/southwestern United States for the CMORPH, CPC U.S.–Mexico, RMORPH, NARR, and Eta Model analyses. As the NERN domain is small relative to the other data sources, it is not shown, but has been examined to compare the intensity of daily rain rates. In all of the representations, the core monsoon region can be seen clearly, extending from about 20° to 30°N along the SMO and the coastal plain in western Mexico. There is reasonable spatial agreement between the five: all show the core region, the extension of the monsoon into southern Arizona, and another peak area of precipitation in northeastern New Mexico.

All five analyses report the highest daily rain rate of the entire NAME domain occurring along the western slopes of the SMO. Predictably, the rain rate of both RMORPH and NARR resemble the CPC U.S.–Mexico daily precipitation analysis, as these rain gauge data are used in the production of both. With the exception of CMORPH, all the analyses produce a daily rain rate in the range of 4–10 mm day−1 for most of the core region. Examination of the CMORPH estimates finds, as expected, a greater (and probably unrealistic) rain rate, with values of 8–22 mm day−1 throughout the core region. The Eta Model forecasts are generally slightly greater than the rates in the central core region from the other estimates, except CMORPH.

While the core region appears clearly in all analyses, the distribution of rainfall intensity varies between the analyses. The region of most intense precipitation (greater than 10 mm day−1) in the RMORPH, CPC U.S.–Mexico, and NARR analyses is in the coastal region of 22°N, just to the south of the opening of the Gulf of California. Examination of precipitation time series of the CPC U.S.–Mexico rain gauge data reveals this is unlikely to be due to a single very high event, as this area records consistently higher precipitation than other regions. The area of maximum rain rate in the Eta Model forecasts (about 8–9 mm day−1) and in CMORPH (18–22 mm day−1) occurs along the western SMO, between about 23° and 27°N.

3. Main features of the diurnal cycle of precipitation and associated fields

Precipitation over the core region of the North American monsoon, along the western slopes of the Sierra Madre Occidental, is considerably greater than in the northern reaches of the monsoon, southern Arizona and New Mexico (see Fig. 1). Therefore, the diurnal cycles of precipitation and some of its associated fields have been analyzed separately in the core region and in the southwestern United States.

a. Structure and evolution of the diurnal cycle of precipitation in the core monsoon region

The diurnal cycle of the core region of the monsoon is inspected using Hovmoeller (time–longitude) diagrams of observed and forecast precipitation, similar to the methods used by Lang et al. (2007) to analyze radar observations. The diurnal cycle is first identified using RMORPH satellite estimates, NARR, and Eta Model forecasts. All figures are presented in local time, and the period of display is from local noon to local noon, to present the continuous diurnal cycle. Figure 2 depicts the diurnal cycle of precipitation in the core region of the monsoon, between 22° and 30°N, 112° and 102°W, for RMORPH satellite estimates (left-hand column), NARR (center column), and Eta Model forecasts (right-hand column). The precipitation has been averaged by 1° latitude bands; Fig. 2 shows every other band.

The area near 22°N, the southern extent of the core monsoon area, was earlier found to have a heavier daily rain rate than the rest of the core monsoon region in NARR and RMORPH. This, and the geographic difference of this area lying to the south of the Gulf of California, suggests an analysis of the diurnal cycle in this area separate from the rest of the core region is in order. In this latitude band, the initiation of precipitation in RMORPH and NARR takes place around 1500–1800 LT, with NARR beginning slightly earlier than RMORPH. For these two analyses, the most intense average precipitation rate is about 1.0–1.2 mm h−1 and in RMORPH occurs between midnight and about 0300 LT and for NARR, about 1800–2400 LT. The Eta Model differs from NARR and RMORPH in that it shows a lower (0.5–0.6 mm h−1) and later (0300–0600 LT) maximum 3-h average rain rate.

In the region of 24°–28°N, all three analyses show the diurnal cycle characterized by the initiation of precipitation, at an average rate of 0.1 mm hr−1, over the crest of the SMO in the afternoon, with NARR beginning earliest (about 1300 LT), RMORPH beginning around 1400–1500 LT, and the Eta Model slightly after 1500 LT. NARR is the most consistent with the NERN rain gauge results of Gochis et al. (2004), which show precipitation in the highest elevation bands beginning to increase at about 1300 LT.

The maximum precipitation rate for these latitude bands occurs along the western slopes of the SMO. RMORPH and NARR show peak rates of precipitation of around 0.7 mm h−1 in this central core region, with the highest rates at 26° to 28°N. For these two analyses, the beginning of maximum precipitation rates take place over the range of 1800–2400 LT, with the NARR maximum precipitation rates starting earlier than RMORPH. Eta forecasts show a peak rate of 1.0–1.2 mm h−1 in this region, slightly displaced to the south. The rainfall diurnal cycle is generally a minimum by about 0300 LT over the coastal plain.

Previous studies have identified a downslope propagation of the diurnal maximum of precipitation over the western slopes of the Sierra Madre Occidental (Gochis et al. 2004; Yarosh et al. 2005; Lang et al. 2007). This behavior is noticeable here in the Hovmoeller diagrams of all three datasets. In this downslope propagation in the core monsoon region, lighter precipitation occurs earlier in the day at higher elevations, while peak rates of precipitation occur later at lower elevations. The diurnal evolution is completed over the coastal plain and the Gulf of California. The downslope shift of precipitation throughout the course of the day is particularly visible in the RMORPH and NARR data, but can be seen as well in the Eta Model output.

An approximation of the speed of the downslope propagation of this signal was obtained by drawing a line along the maximum values in each Hovmoeller diagram (see, e.g., Fig. 2b, central panel); the slope of this line represents the speed of propagation. The slope of this line (speed) has been calculated for each degree of latitude. The results of this analysis are presented in Fig. 3. RMORPH, NARR, and the Eta Model all show speeds of downslope propagation near zero at 20°N. Maximum speeds occur between 22° and 28°N, the area lying along the main body of the Gulf of California. NARR shows a speed over 6 m s−1 in most of the core region, with a peak speed of nearly 8 m s−1, in line with the results of Lang et al. (2007). RMORPH shows slower peak speeds, about 5 m s−1. While NARR and RMORPH are produced using the same daily precipitation analysis, different disaggregation methods are used in their production and differences appear between their diurnal cycles, such as the speed of the westward shift. The Eta Model shows a yet slower speed of westward shift, around 2.5 m s−1.

b. Associated fields of the diurnal cycle in the core region

The evolution of the diurnal cycle of precipitation and the downslope propagation of the precipitation maximum observed in the diurnal cycle have been further assessed by the combined analysis of several associated fields. These include convective available potential energy (CAPE), convective inhibition (CIN), and the moisture flux convergence. An additional factor, evaporation, has also been analyzed, in the interest of understanding the sources of atmospheric moisture. It is hypothesized that evaluating the strength of these associations will provide insight into the diurnal cycle of precipitation.

CAPE is the maximum energy available to an ascending air parcel, while CIN is the amount of energy required to lift a parcel of air from the surface to the level of free convection. In the core region, CAPE increases from near zero before monsoon onset in early July to values well over 1000 J kg−1 and considerably larger instant values during its mature phase (Barlow et al. 1998). Vertically integrated moisture flux convergence (MFC) represents dynamically transported moisture, while evaporation provides a local source of precipitable water in the atmosphere. The Gulf of California presents a large source of moisture for the northwestern Mexico region, and previous studies of the seasonal evolution of moisture flux (e.g., Higgins et al. 1997; Barlow et al. 1998) have found close agreement between the onset of the monsoon and an increase in MFC in the core monsoon region. All of these fields, while interrelated, may be individually analyzed to assess the strength of association between their diurnal cycles and that of precipitation.

Figure 4 depicts Hovmoeller diagrams of the diurnal cycle of the previous terms. The July and August 2004 diurnal cycle of CAPE anomalies (left column) and of MFC for the core monsoon region were obtained from NARR. The daily average of CAPE, which ranges from about 3000 J kg−1 over the Gulf of California to about 200 J kg−1 over the coastal plain and western SMO, has been removed to highlight the diurnal variation. As in Fig. 2, to present the continuous diurnal cycle, the figures are from local noon to local noon, and precipitation contours from NARR have been overlain.

In the afternoon and evening, a maximum in CAPE occurs over the western slopes of the SMO and the coastal plain, with a minimum over the Gulf of California. CAPE builds up over the land throughout the afternoon, and the maximum over land occurs around 1800 LT, shortly after the warmest part of the day. Between midnight and 0900 LT, when the waters of the Gulf of California are warmer than the land, higher CAPE forms over the Gulf of California, and a minimum appears over the land. The middle three panels of the left side of Fig. 4 show latitudes 24°–28°N, the area along the Gulf of California, where the diurnal cycle of CAPE has two maxima, one in the afternoon/evening over land and one over the Gulf after midnight. This second maximum represents a much smaller variation from the average, as the daily average of CAPE over the Gulf of California is much greater than that over the land. The bottom panel, 22°N, is south of the Gulf of California, and there is no postmidnight second maximum, rather CAPE peaks over the land in the evening and decreases steadily to the west throughout the rest of the night.

Figure 4 also shows midafternoon moisture flux divergence over the Gulf of California and MFC over the western slopes of the Sierra Madre Occidental, similar to the results of Berbery (2001). The core region experiences afternoon upslope flow (Stensrud et al. 1995; Johnson et al. 2007) that brings moisture in from the Gulf of California and converges against the western slopes of the SMO. MFC at all latitudes peaks around 1800–2100 LT. Figure 4 shows postmidnight MFC along the coastal plain: by midnight, upslope flow has dissipated and southerly flow has developed along the Gulf of California and the coastal plain (Douglas 1995), which, coupled with downslope flow off the SMO, leads to this pattern of MFC. The strongest MFC appears to precede the peak precipitation by about 3 h. MFC is weakest at 30°N, where the slope of the SMO is the most gradual and the distance to the Gulf of California the greatest. It is strongest at 22°N, where the coastal plain is very narrow and the slope of the SMO is steep. The diurnal pattern of MFC (Fig. 4) shows evidence of a westward propagation similar to that of precipitation, with the MFC diurnal maximum occurring slightly before the precipitation maximum, and the area of MFC moving off to the west throughout the day, diminishing in magnitude. In the region of 24°–28°N, similar to CAPE, a smaller, secondary maximum in both precipitation and MFC can also be noted over the Gulf of California in the postmidnight hours.

To compare the timing of precipitation and the associated fields in the core region, the land-only mean diurnal cycle area average of precipitation, MFC, evaporation, CAPE, and CIN are shown in Fig. 5. NARR results are complemented with Eta Model forecast computations. The area between 20° and 30°N and 111° to 104°W is shown. Intensities in the core region have been damped by the area averaging, as the core region is a narrow strip along the western slopes of the SMO and the coastal plain. NARR (Fig. 5a) shows MFC and precipitation both peaking at about 2100 LT. CAPE peaks about 3 h before peak precipitation and coincides with the minimum in CIN. Evaporation peaks during the warmest part of the day, about 6 h before peak precipitation. Compared to the Eta Model results, NARR shows higher evaporation and CAPE and lower MFC for the core region. The higher evaporation may be related to the higher sea surface temperatures along the Gulf of California that NARR employed (see Mesinger et al. 2006).

c. The diurnal cycle of precipitation and associated fields in southern Arizona and New Mexico

While the North American monsoon is very important in the southwestern United States, supplying up to 60% of the annual precipitation (Douglas et al. 1993), the daily precipitation rain rate is only on the order of 1–2 mm day−1, contrasted with the more than 8 mm day−1 in the core region. The southwestern United States is subject to neither land–water contrasts nor to the sloping of the SMO, and the average precipitation diurnal cycle does not exhibit a westward shift.

Area averages of the region between 32° and 36°N, 113°–104°W have been used to analyze the diurnal cycle of precipitation and its associated fields in southern Arizona and New Mexico (Fig. 6). As stated earlier, Mo et al. (2005) found that NARR systematically overestimates the Gulf of California LLJ water transport, producing too-strong winds at the surface and aloft. As a result of this bias, the corresponding MFC is much stronger in NARR than in the Eta Model forecasts. For this reason, alongside NARR results, the Eta Model forecasts are employed as a more realistic representation of the moisture fluxes.

The diurnal cycle in NARR (Fig. 6a) shows a minimum in all factors (precipitation, MFC, evaporation, and CAPE and a maximum in CIN magnitude) at about 0600–0900 LT, soon after the diurnal temperature minimum. As the surface temperature warms, evaporation and CAPE increase to their diurnal maxima at about 1500 LT. The precipitation rate peaks at about 1800 LT, and MFC also has a peak at this time, similar to the findings of Anderson and Kanamaru (2005). MFC decreases after 1800 LT, until about midnight, and peaks again at about 0300 LT. This second peak may be the result of the nocturnal LLJ over the Gulf of California, which generally peaks at about 0100 LT (Douglas 1995; Douglas et al. 1998; Mo et al. 2005). This LLJ brings moisture north from the Gulf into the southwestern United States (Douglas 1995).

For the reasons already discussed, the NARR diurnal cycle, with strong convergence around 0300 LT, then values near zero during the morning and early afternoon, and convergence again in the late afternoon, may be biased. The Eta Model, on the other hand, shows weak convergence in the morning and relatively strong divergence in the afternoon and evening. MFC peaks about 12 h before precipitation peaks, suggesting a weaker association between these two fields than between CAPE and precipitation, as CAPE peaks 3 to 6 h before precipitation. In summary, precipitation, CAPE, CIN, and evaporation are similar in timing and magnitude for both NARR and the Eta Model forecasts, and the major difference between the two is found in MFC. However, to better understand the links between precipitation and other variables, it is necessary to separate surge and no-surge cases, as will be discussed next.

d. Moist surges and precipitation in the core region and the southwestern United States

Thus far the discussion has been on time-mean conditions over the southwestern United States. However, northward surges of cool, moist air along the Gulf play a critical role in the precipitation mechanisms in the southwestern United States (Douglas and Leal 2003; Higgins et al. 2004). Abrupt changes in the low-level moisture flux have been identified as a distinct component of moisture surges (Douglas and Leal 2003), and surges have been found to account for a majority of the precipitation in Arizona in model simulations (Berbery and Fox-Rabinovitz 2003) and in observations (Higgins et al. 2004). Surges are usually produced by the passage of an easterly wave; stronger surges are formed when a midlatitude disturbance occurs in conjunction with the passage of the easterly wave (Stensrud et al. 1997).

Moisture surges were identified here as when the 950-hPa meridional moisture flux of the Eta Model forecasts in the northern Gulf of California equals or exceeds the mean plus 50% of the standard deviation, following the method of Berbery and Fox-Rabinovitz (2003). All other cases are deemed to be nonsurge times. Due to the aforementioned overestimation of meridional moisture transport in NARR, Eta Model forecasts were used for this calculation. Many surge indices have been proposed (e.g., Fuller and Stensrud 2000; Higgins et al. 2004), and this definition simply takes into account strong moisture flux cases, which have traveled to the north of the Gulf of California. Tests with the meridional moisture flux threshold computed at points in the southern and central Gulf of California (not shown) produced a greater number of high moisture flux periods, but these were of shorter duration than the four periods seen when the northern Gulf is analyzed.

Figure 7 presents the percentage of total July–August 2004 precipitation that occurred during surge events, estimated from the Eta Model forecast and RMORPH data. The total amount of precipitation that fell during surges was compared to the total precipitation of the entire 2-month period to produce this figure. NARR precipitation is similar to RMORPH and thus for simplicity is not shown. In most of the core region, both the Eta Model and RMORPH show only a moderate dependence on surges, with generally between 30% and 50% of precipitation occurring during surges, similar to the findings of Douglas and Leal (2003). In the southwestern United States, both the Eta Model and RMORPH show high percentages of precipitation occurring during surges (upward of 70%), also in line with the results of Douglas and Leal (2003). During surges, both Eta forecasts and RMORPH show high percentages of precipitation occurring in Southern California and Arizona. The major difference between the two datasets is in a small area of the western coastline of the northern portion of the Gulf of California, where RMORPH shows high percentages of precipitation occurring during surges. This percentage decreases to the east, with a minimum over the northern peaks of the SMO. Conversely, the Eta Model shows lower percentages along the coast, increasing to the east.

The meridional moisture flux in the northern Gulf of California from the Eta Model and NARR is shown in Fig. 8a; as expected, NARR is generally (although not always) stronger than the Eta Model, likely due to NARR’s bias in meridional winds of the Gulf of California LLJ. Figures 8b and 8c illustrate the time series of precipitation for both the Eta Model forecasts and for NARR for the period of 10 July through 31 August 2004; surge times are marked with shaded areas. Two areas are shown, area-averaged southern Arizona and New Mexico, between 32° and 36°N, 113° and 104°W, and the core monsoon region, between 20° and 30°N, from 111° to 104°W. Both Eta Model output and NARR are displayed as 8-point running means of 3-h data to remove the diurnal cycle of the time series. In this analysis, four surge periods are observed, covering approximately 12–16 and 23–27 July and 10–17 and 23–24 August. Tropical Storm Blas, occurring 11–15 July 2004, appears as a strong meridional moisture flux and precipitation event and produced a strong surge (Higgins et al. 2006). During the period 22–26 July, widespread thunderstorms were observed in the tier I NAME region (see http://catalog.eol.ucar.edu/name), and this period featured a strong surge as well (Johnson et al. 2007).

In the core region, NARR tends to show a more uniform distribution of precipitation, giving less relevance to surge/no-surge cases. On the other hand, Eta Model forecasts are more sensitive to surge/no-surge cases, generally showing more precipitation during surges than NARR, but less precipitation than NARR during nonsurge times. While large amounts of precipitation occur in the core region during the two July surge periods, precipitation during the major August surge is not particularly strong, and several smaller but distinct periods of high precipitation occurred on days without a surge.

In southern Arizona and New Mexico, maximum precipitation occurs during surge times, as represented in both the Eta Model forecasts and NARR. Eta Model precipitation forecasts are generally slightly higher than NARR in the Arizona and New Mexico area average, especially during surge times. Three of the four strong precipitation events in Arizona coincided with a moisture surge.

e. Dependence of the diurnal cycle in the southwestern United States on surges

Given the importance of surges for precipitation in the southwestern United States, we now separately examine the diurnal cycle of precipitation and of its associated fields in southern Arizona and New Mexico for surge and nonsurge times, as defined earlier in this section. To check how representative was the summer of 2004, with respect to other years, a 4-yr NARR climatology computed over July, August, and September of 2002 through 2005 is also presented for surge and nonsurge times. In this case, in the absence of the Eta Model forecasts to compute the moisture flux index that defines the surge and no-surge cases, the operational NCEP Eta Data Assimilation System (EDAS) analysis was employed.

First, the surge and nonsurge daily average for precipitation, evaporation, MFC, CAPE, and CIN are shown in Table 1 for NARR (including the 4-yr climatology and the NAME 2004 field campaign season), and for the Eta Model. Looking at this table, some generalizations can be drawn from comparing the results for surges to the results of nonsurge times, for all three datasets. The daily average for precipitation is about twice as high during surges than in no-surge cases, while evaporation has a very similar daily average during both surges and nonsurge times. CAPE is also nearly doubled during surges, unlike CIN that exhibits a small increase in magnitude. The field that shows the greatest difference between surges and nonsurge times is MFC, as expected. The daily average magnitude of MFC is considerably higher during surges. Furthermore, examination of the 2004 season from NARR finds that the daily averages for both surges and nonsurges are very similar to their counterparts in the 4-yr climatology of NARR, indicating that 2004 is a representative season.

As in Fig. 6, Fig. 9 shows the area-averaged diurnal cycle of precipitation, evaporation, MFC, CAPE, and CIN for NARR and for the Eta Model. Figures 9a and 9b show that the 4-yr NARR climatology is generally similar to the patterns during the NAME 2004 field campaign of July and August, that is, Figs. 9c and 9d. According to Figs. 9c and 9d, the NARR precipitation rate peaks about 3 h earlier in the day during surges (1800 LT) than in the nonsurge cases (2100 LT) and at higher intensities. On the other hand, evaporation exhibits a very similar diurnal evolution and daily average (see Table 1) during surge and nonsurge times. CIN has a higher peak in the morning hours during surges but drops off to a minimum by 1800 LT that is approximately the same magnitude as the minimum seen in the nonsurge case. CAPE shows higher intensities throughout its diurnal cycle.

Table 1 showed an important increase in MFC during surges both in NARR and Eta Model forecasts. Changes are found in the evolution of the diurnal cycle as well, as seen in Fig. 9. In the case of NARR, three MFC peaks are noticed: at 0300, 0900, and 1800 LT. When each of the 4 yr used for the surge climatology (Fig. 9a) were analyzed individually (not shown) similar multipeak patterns were found in the MFC diurnal cycles, with the peaks occurring at various times throughout the four cycles. On the other hand, when the nonsurge MFC diurnal cycles were analyzed, all 4 yr exhibited very similar patterns to each other and to the 4-yr average. While the rather unclear, multipeak pattern seen in the surge cases may be attributed to the effect of individual surge cases (four surge cases in the 2004 season, for example), it is also possibly due to NARR’s inaccurate representation of the meridional moisture fluxes north of the Gulf of California. Notice that the Eta Model forecasts have a seemingly more consistent behavior: in the no-surge case (Fig. 9f) the MFC remains close to zero and even becomes negative (divergence) during part of the day. Therefore, it does not seem to have a strong association with the development of precipitation. However, during surges, MFC develops a better-defined diurnal cycle with a maximum preceding precipitation by about 6 h, suggesting a stronger association between this field and precipitation during surges.

4. Summary and conclusions

The purpose of this study has been to analyze the structure of the diurnal cycle of precipitation and associated fields during the North American Monsoon Experiment (NAME) field campaign of July and August 2004. A proper representation of physical mechanisms, such as the diurnal cycle, is essential to understanding and predicting warm-season precipitation in North America. The analysis was conducted using RMORPH satellite estimates, derived from the disaggregation of the Climate Prediction Center (CPC) U.S.–Mexico daily rain gauge analysis via CMORPH satellite estimates, model forecasts from the workstation version of NCEP’s Eta Model, and reanalysis data from the North American Regional Reanalysis (NARR). The CPC U.S.–Mexico rain gauge and NAME Event Rain Gauge Network (NERN) analyses were also examined for comparison purposes.

The tier I NAME region was examined both as a whole and as two separate regions, the core monsoon region along the western side of the Sierra Madre Occidental (SMO) and the southwestern United States. Due to the importance of northward surges of moist air from the Gulf of California into the southwestern United States, the effect of surges on the diurnal cycle was examined for this region as well. This study looked at the precipitation diurnal cycle and several associated fields, including evaporation, the convective available potential energy, and convective inhibition, as well as the moisture flux convergence.

NARR, Eta Model forecasts, and RMORPH satellite estimates produce very similar patterns and daily average precipitation and correspond well to rain gauge observations. On the other hand, CMORPH satellite precipitation estimates systematically produce high daily rain rates, up to 3 times as high as the other sources.

Comparison of RMORPH estimates, NARR, and Eta Model forecasts revealed similar diurnal cycles in the core monsoon region. The 24-h cycle of precipitation begins in the afternoon along the crest of the SMO, reaching peak rates around 2100 LT along the western slopes of the SMO, with precipitation ending in the early morning hours over the coastal plain. Consistent with previous studies, a westward shift of precipitation is seen in the core region: the precipitation cycle initiates and completes earlier at higher topographical elevations and later toward the west. This westward shift moves at different speeds over the range of the core monsoon region depending on the latitude, with the highest speeds occurring in the region of 24°N–28°N, where the Sierra Madre Occidental reaches the highest elevations.

Although NARR and RMORPH start with the same daily rain gauge database, the two use different methods to disaggregate the daily precipitation, and important differences can be noticed in their representation of the diurnal cycle. The NARR precipitation diurnal cycle generally begins earlier in the day than RMORPH, and NARR shows a faster westward shift of the precipitation maximum throughout the day than RMORPH. As stated, these differences may be in part related to the difference in resolution of the disaggregating methods: according to Mesinger et al. (2006) the subdiurnal variability over Mexico is obtained from the NCEP–DOE Global Reanalysis 2. The diurnal cycle of RMORPH results from the satellite estimates derived by Joyce et al. (2004) for the 0.25° × 0.25° CMORPH dataset. The Eta Model precipitation diurnal cycle generally starts later than the other two and exhibits a slower, less distinct westward shift.

Several fields associated with the diurnal cycle of precipitation in the core monsoon region were examined using information from NARR, including CAPE and CIN and the vertically integrated MFC. MFC exhibits a westward shift similar to that of precipitation, with the maximum occurring in the late afternoon, shortly before the precipitation maximum, and moving to the west throughout the evening. The CAPE diurnal cycle has two maxima, one in the afternoon over the coastal plain, and one in the early morning over the Gulf of California.

Northward moisture surges along the Gulf of California produce a noticeable increase in precipitation in the northern edge of the monsoon region, specifically the southwestern United States. Generally, the diurnal cycle in nonsurge cases resembles the overall average, probably reflecting the greater frequency of no-surge cases. Surges produce higher precipitation and an altered diurnal cycle in some of the associated fields, especially MFC. In the Eta Model, the moisture flux has a poorly defined diurnal cycle in nonsurge cases and remains divergent throughout the day, but during surges it becomes convergent and acquires a well-defined evolution with a peak about 6 h before peak precipitation, suggesting that MFC has a greater role in precipitation during surges. Also, during surges, as seen in both NARR and the Eta Model, CAPE and CIN have modifications only to the amplitude of their cycles. As also seen in previous studies, NARR appears to overestimate the meridional transport of water in this region.

Acknowledgments

The authors thank Dr. Bruce Anderson, Dr. Steve Nesbitt, and one anonymous reviewer for their thoughtful comments and suggestions that improved our manuscript. We are thankful to Drs. Wei Shi and John Janowiak for supplying the RMORPH dataset and Dr. Wayne Higgins for his suggestions. We thank Dr. David Gochis for supplying the NERN dataset and for his comments. This work was supported by NOAA Grants NA17EC1483 and NA04OAR4310164.

REFERENCES

  • Adams, D. K., , and A. C. Comrie, 1997: The North American monsoon. Bull. Amer. Meteor. Soc., 78 , 21972213.

  • Anderson, B. T., , and H. Kanamaru, 2005: The diurnal cycle of the summertime atmospheric hydrologic cycle over the southwestern United States. J. Hydrometeor., 6 , 219228.

    • Search Google Scholar
    • Export Citation
  • Anderson, B. T., , J. O. Roads, , and S-C. Chen, 2000: Large-scale forcing of summertime monsoon surges over the Gulf of California and the southwestern United States. J. Geophys. Res., 105 , 2445524467.

    • Search Google Scholar
    • Export Citation
  • Anderson, B. T., , J. O. Roads, , S-C. Chen, , and H-M. H. Juang, 2001: Model dynamics of summertime low-level jets over northwestern Mexico. J. Geophys. Res., 106 , 34013413.

    • Search Google Scholar
    • Export Citation
  • Barlow, M., , S. Nigam, , and E. H. Berbery, 1998: Evolution of the North American monsoon system. J. Climate, 11 , 22382257.

  • Berbery, E. H., 2001: Mesoscale moisture analysis of the North American monsoon. J. Climate, 14 , 121137.

  • Berbery, E. H., , and M. S. Fox-Rabinovitz, 2003: Multiscale diagnosis of the North American monsoon system using a variable-resolution GCM. J. Climate, 16 , 19291947.

    • Search Google Scholar
    • Export Citation
  • Bordoni, S., , P. E. Ciesielski, , R. H. Johnson, , B. D. McNoldy, , and B. Stevens, 2004: The low-level circulation of the North American Monsoon as revealed by QuikSCAT. Geophys. Res. Lett., 31 .L10109, doi:10.1029/2004GL020009.

    • Search Google Scholar
    • Export Citation
  • Douglas, M. W., 1995: The summertime low-level jet over the Gulf of California. Mon. Wea. Rev., 123 , 23342348.

  • Douglas, M. W., , and J. C. Leal, 2003: Summertime surges over the Gulf of California: Aspects of their climatology, mean structure, and evolution from radiosonde, NCEP reanalysis, and rainfall data. Wea. Forecasting, 18 , 5574.

    • Search Google Scholar
    • Export Citation
  • Douglas, M. W., , R. A. Maddox, , and K. Howard, 1993: The Mexican monsoon. J. Climate, 6 , 16651677.

  • Douglas, M. W., , A. Valdez-Manzanilla, , and R. G. Cueto, 1998: Diurnal variation and horizontal extent of the low-level jet over the northern Gulf of California. Mon. Wea. Rev., 126 , 20172025.

    • Search Google Scholar
    • Export Citation
  • Farfán, L. M., , and J. A. Zehnder, 1994: Moving and stationary mesoscale convective systems over northwest Mexico during the Southwest Area Monsoon Project. Wea. Forecasting, 9 , 630639.

    • Search Google Scholar
    • Export Citation
  • Fawcett, P. J., , J. R. Stalker, , and D. S. Gutzler, 2002: Multistage moisture transport into the interior of northern Mexico during the North American summer monsoon. Geophys. Res. Lett., 29 .2094, doi:10.1029/2002GL015693.

    • Search Google Scholar
    • Export Citation
  • Fuller, R. D., , and D. J. Stensrud, 2000: The relationship between tropical easterly waves and surges over the Gulf of California during the North American monsoon. Mon. Wea. Rev., 128 , 29832989.

    • Search Google Scholar
    • Export Citation
  • Gochis, D. J., , J-C. Leal, , C. J. Watts, , W. J. Shuttleworth, , and J. Gartuza-Payan, 2003: Preliminary diagnostics from a new event-based precipitation monitoring system in support of the North American Monsoon Experiment. J. Hydrometeor., 4 , 974981.

    • Search Google Scholar
    • Export Citation
  • Gochis, D. J., , A. Jimenez, , C. J. Watts, , J. Garatuza-Payan, , and W. J. Shuttleworth, 2004: Analysis of 2002 and 2003 warm-season precipitation from the North American Monsoon Experiment event rain gauge network. Mon. Wea. Rev., 132 , 29382953.

    • Search Google Scholar
    • Export Citation
  • Hales Jr., J. E., 1972: Surges of maritime tropical air northward over the Gulf of California. Mon. Wea. Rev., 100 , 298306.

  • Hales Jr., J. E., 1974: Southwestern United States summer monsoon source—Gulf of Mexico or Pacific Ocean? J. Appl. Meteor., 13 , 331342.

    • Search Google Scholar
    • Export Citation
  • Higgins, R. W., , and W. Shi, 2005: Relationships between Gulf of California moisture surges and tropical cyclones in the eastern Pacific basin. J. Climate, 18 , 46014620.

    • Search Google Scholar
    • Export Citation
  • Higgins, R. W., , Y. Yao, , and X. L. Wang, 1997: Influence of the North American monsoon system on the U.S. summer precipitation regime. J. Climate, 10 , 26002621.

    • Search Google Scholar
    • Export Citation
  • Higgins, R. W., , W. Shi, , E. Yarosh, , and R. Joyce, 2000: Improved United States Precipitation Quality Control System and Analysis. NCEP/Climate Prediction Center Atlas 7, 40 pp. [Available online at http://www.cpc.ncep.noaa.gov/research_papers/ncep_cpc_atlas/7/index.html.].

    • Search Google Scholar
    • Export Citation
  • Higgins, R. W., , W. Shi, , and C. Hain, 2004: Relationships between Gulf of California moisture surges and precipitation in the southwestern United States. J. Climate, 17 , 29832995.

    • Search Google Scholar
    • Export Citation
  • Higgins, R. W., and Coauthors, 2006: The NAME 2004 field campaign and modeling strategy. Bull. Amer. Meteor. Soc., 87 , 7994.

  • Howard, K. W., , and R. A. Maddox, 1988: Mexican mesoscale convective systems—A satellite perspective. Preprints, Third Int. American and Mexican Congress of Meteorology, Mexico City, Mexico, Mexican Meteorological Organization, 404–408.

  • Janowiak, J. E., , P. Xie, , R. Joyce, , M. Chen, , and Y. Yarosh, 2004: Validation of daily satellite precipitation estimates over the U.S. Proc. 29th Annual Climate Diagnostics and Prediction Workshop, Madison, WI, NOAA. [Available online at http://www.cpc.noaa.gov/products/outreach/proceedings/cdw29_proceedings/CDW29.proceedings.shtml.].

    • Search Google Scholar
    • Export Citation
  • Janowiak, J. E., , V. J. Dagostaro, , V. E. Kousky, , and R. J. Joyce, 2007: An examination of precipitation in observations and model forecasts during NAME with emphasis on the diurnal cycle. J. Climate, 20 , 16801692.

    • Search Google Scholar
    • Export Citation
  • Johnson, R. H., , P. E. Ciesielski, , B. D. McNoldy, , P. J. Rogers, , and R. K. Taft, 2007: Multiscale variability of the flow during the North American Monsoon Experiment. J. Climate, 20 , 16281648.

    • Search Google Scholar
    • Export Citation
  • Joyce, R. J., , J. E. Janowiak, , P. A. Arkin, , and P. Xie, 2004: CMORPH: A method that produces global precipitation estimates from passive microwave and infrared data at high spatial and temporal resolution. J. Hydrometeor., 5 , 487503.

    • Search Google Scholar
    • Export Citation
  • Kistler, R., and Coauthors, 2001: The NCEP–NCAR 50-Year Reanalysis: Monthly means CD-ROM and documentation. Bull. Amer. Meteor. Soc., 82 , 247267.

    • Search Google Scholar
    • Export Citation
  • Lang, T. J., , D. A. Ahijevych, , S. W. Nesbitt, , R. E. Carbone, , S. A. Rutledge, , and R. Cifelli, 2007: RADAR-observed characteristics of precipitating systems during NAME 2004. J. Climate, 20 , 17131733.

    • Search Google Scholar
    • Export Citation
  • McCollum, J. R., , W. F. Krajewski, , R. R. Ferraro, , and M. B. Ba, 2002: Evaluation of biases of satellite rainfall estimation algorithms over the continental United States. J. Appl. Meteor., 41 , 10651080.

    • Search Google Scholar
    • Export Citation
  • Mesinger, F., and Coauthors, 2006: North American Regional Reanalysis. Bull. Amer. Meteor. Soc., 87 , 343360.

  • Mo, K., , M. Chelliah, , M. L. Carrera, , R. W. Higgins, , and W. Ebisuzaki, 2005: Atmospheric moisture transport over the United States and Mexico as evaluated in the NCEP regional reanalysis. J. Hydrometeor., 6 , 710728.

    • Search Google Scholar
    • Export Citation
  • Rosenfeld, D., , and Y. Mintz, 1988: Evaporation of rain falling from convective cloud as derived from radar measurements. J. Appl. Meteor., 27 , 209215.

    • Search Google Scholar
    • Export Citation
  • Shafran, P., , J. Woollen, , W. Ebisuzaki, , W. Shi, , Y. Fan, , R. Grumbine, , and M. Fennessy, 2004: Observational data used for assimilation in the NCEP North American regional reanalysis. Preprints, 14th Conf. on Applied Climatology, Seattle, WA, Amer. Meteor. Soc., 1.4.

  • Smith, W. P., , and R. L. Gall, 1989: Tropical squall lines of the Arizona monsoon. Mon. Wea. Rev., 117 , 15531569.

  • Stensrud, D. J., , R. L. Gall, , S. L. Mullen, , and K. W. Howard, 1995: Model climatology of the Mexican monsoon. J. Climate, 8 , 17751794.

    • Search Google Scholar
    • Export Citation
  • Stensrud, D. J., , R. L. Gall, , and M. K. Nordquist, 1997: Surges over the Gulf of California during the Mexican monsoon. Mon. Wea. Rev., 125 , 417437.

    • Search Google Scholar
    • Export Citation
  • Yarosh, Y., , P. Xie, , M. Chen, , R. Joyce, , J. E. Janowiak, , and P. A. Arkin, 2005: Diurnal cycle in the North American monsoon. Bull. Amer. Meteor. Soc., 86 , 2628.

    • Search Google Scholar
    • Export Citation

Fig. 1.
Fig. 1.

Daily average precipitation (mm day−1) over the North American monsoon region from July and August 2004: (a) CMORPH satellite estimates, (b) CPC U.S.–Mexico daily rain gauge analysis, (c) RMORPH satellite estimates, (d) North American Regional Reanalysis, and (e) Eta Model forecasts. The core monsoon region is clearly visible along the western slopes of the Sierra Madre Occidental between approximately 20° and 30°N.

Citation: Journal of Climate 21, 4; 10.1175/2007JCLI1642.1

Fig. 2.
Fig. 2.

Hovmoeller diagrams of the diurnal cycle of precipitation in the core monsoon region, averaged into 1° latitude bands. This figure displays every other latitude. The diurnal cycle is shown from local noon to local noon to show the complete cycle; the contour interval is 0.1 mm h−1; (a) RMORPH, (b) NARR, and (c) Eta Model forecasts.

Citation: Journal of Climate 21, 4; 10.1175/2007JCLI1642.1

Fig. 3.
Fig. 3.

Speed of westward shift of precipitation over the core monsoon region, as obtained from the slope of the Hovmoeller diagrams of precipitation from RMORPH, NARR, and the Eta Model.

Citation: Journal of Climate 21, 4; 10.1175/2007JCLI1642.1

Fig. 4.
Fig. 4.

Hovmoeller diagrams of the diurnal cycle of (left) CAPE and (right) the vertically integrated moisture flux convergence from NARR. Precipitation contours have been overlain. The CAPE contour interval is 100 J kg−1, MFC contour interval is 0.2 kg m−2, and the precipitation contour interval is 0.1 mm h−1. To highlight the diurnal variation, the daily average of CAPE, which ranges from about 3000 J kg−1 over the Gulf of California to about 200 J kg−1 over the coastal plain and western SMO, has been removed. Both CAPE and the moisture flux convergence have been smoothed by the application of a 9-point smoothing function.

Citation: Journal of Climate 21, 4; 10.1175/2007JCLI1642.1

Fig. 5.
Fig. 5.

Land-only area average of several elements in the core monsoon region from (a) NARR and (b) the Eta Model for the 24-h period beginning at local midnight. The upper panels depict precipitation (solid line, mm h−1), moisture flux convergence (long dash, mm day−1), and evaporation (dotted line, mm h−1). The middle panels are CAPE (J kg−1), and the bottom panels are CIN (J kg−1).

Citation: Journal of Climate 21, 4; 10.1175/2007JCLI1642.1

Fig. 6.
Fig. 6.

Same as in Fig. 5 but for the southwestern United States.

Citation: Journal of Climate 21, 4; 10.1175/2007JCLI1642.1

Fig. 7.
Fig. 7.

Percentage of precipitation during surge events for (left) Eta Model forecasts and (right) RMORPH satellite estimates. Shaded areas represent greater than 50% of total precipitation occurring during surge events; white areas are less than 50%. The total amount of precipitation that fell during surges was compared to the total precipitation of the entire 2-month period to produce this figure.

Citation: Journal of Climate 21, 4; 10.1175/2007JCLI1642.1

Fig. 8.
Fig. 8.

(a) Meridional moisture flux at 30°N, 114°W, the northern Gulf of California; (b) area-averaged precipitation for the southwestern United States; and (c) the core North American monsoon region (bottom) for 10 Jul–31 Aug 2005. Eta Model forecasts (solid) and NARR satellite estimates (dashed); surge times are indicated by shaded regions.

Citation: Journal of Climate 21, 4; 10.1175/2007JCLI1642.1

Fig. 9.
Fig. 9.

As in Fig. 6 but for the southwestern United States, divided into surge and nonsurge times. (a), (b) The 4-yr climatology from NARR of this region divided into surge and nonsurge. (c), (d) The 2004 NAME field campaign period obtained from NARR. (e), (f) The 2004 NAME field campaign period, as represented in the Eta Model. The upper panels depict precipitation (solid line, mm h−1), moisture flux convergence (long dashed line, mm day−1), and evaporation (dotted line, mm h−1). The middle panels are CAPE (J kg−1), and the bottom panels are CIN (J kg−1).

Citation: Journal of Climate 21, 4; 10.1175/2007JCLI1642.1

Table 1.

Daily average values for surge and nonsurge conditions for the 4-yr NARR average and the NARR and the Eta Model 2004 season.

Table 1.
Save