Review of Variability in the North American Monsoon
David K. Adams
Dept. of Geography and Regional Development
University of Arizona
The North American monsoon (NA monsoon), variously known as the Southwest United States monsoon, the Mexican monsoon, or the Arizona monsoon, is experienced as a pronounced increase in rainfall from an extremely dry June to a rainy July over large areas of the southwestern United States and northwestern Mexico. These summer rains typically last until mid-September when a drier regime is reestablished over the region. Geographically, the NA monsoon precipitation region is centered over the Sierra Madre Occidental in the Mexican states of Sinaloa, Durango, Sonora and Chihuahua. The regime extends northward into the Arizona, New Mexico and Colorado. Typically, the NA Monsoon region is defined by sites that recieve at least 50% of its annual precipitation in July,
August and September (Figure 1). The NA monsoon has been studied during the last century and quite intensively during the last decade (Adams and Comrie, in press). One of principal motivations for much of the scientific investigation of the NA Monsoon phenomenon has been to understand its variability, not only seasonally, but also interannually. Apart from the purely scientific challenge of understanding of monsoon dynamics, variability in summer rainfall is of pratical concern for watershed managers, ranchers and planners of southwestern North America.
NA Monsoon Variability
Unlike the Pacific region of the United States and the west coast of Baja California, precipitation in the NA monsoon is not associated with large-scale mid-latitude cyclones, but with convective activity (thunderstorms) which has very different spatial/temporal distribution characteristics. The difficulty in understanding the variability of summertime convective activity in the southwestern U.S. and northwestern Mexico results from the complex interactions between atmospheric circulation features at both the synoptic (100 to 1000 km spatially, 1 day to 1 week, temporally) and mesoscale (several kms to 100 km, hours to one day temporally) and the extremely varied topography. The larger-scale atmospheric motions may control the distribution of water vapor and the general stability or instability (that is, the tendency to form storms) in the atmosphere; nevertheless, local topographic effects are critical to the geographic and even temporal distribution of convective activity. The fact that there is often little clear-cut relationship between synoptic-scale circulation and the outbreak of thunderstorm activity has complicated modeling and forecasting efforts for this region (Dunn and Horel 1994).
Spatial Variability at the Regional Scale
It has been shown that the geographic influence of NA monsoon rains is primarily centered in the western foothills of the Sierra Madre Occidental in northwestern Mexico, specifically in the Mexican states of Nayarit, Sinaloa, and Sonora (Douglas et al. 1993) (Figure 2). This part of Mexico receives up to 70 percent of its annual rainfall in the months of July, August and September. The marked July through September maximum decreases rapidly as one moves northward through Sonora approaching the U.S. border. The NA monsoon precipitation regime becomes less notable approaching the western flank of the Sierra Madre Oriental where there is a general decrease in July and August rainfall (discussed in the following section). Similarly, westward across the Gulf of California along the Baja peninsula, NA monsoon rains diminish rapidly as the influence of the cool waters of the Pacific and increased atmospheric subsidence (sinking air warms, leading to greater atmospheric stability) under the Pacific sub-tropical high stifle convective activity. Southern and eastern Arizona and much of New Mexico lie climatologically on the fringes of this principally Mexican phenomenon. As a result, summer precipitation becomes much more variable and its distribution becomes even more influenced by the topography of these states -- the higher elevations receiving a much greater occurrence of thunderstorm activity and precipitation. The spatial distribution of precipitation in the interior regions is directly related to the penetration of maritime tropical air (Hales 1972). In addition, Carleton (1986) feels that synoptic-scale controls are equally important for the regional variability.
Apart from the spatial variability that is directly related to the topography, there is, in fact, much intraseasonal variability in terms of the intensity and areal extent of NA monsoon rainfall particularly in the southwestern U.S.. Northwestern Mexico, specifically the highlands of eastern Sinaloa and Sonora, experience nearly daily shower activity during the entire monsoon season. However, interseasonal precipitation variability in this region may be related to the frequency of mesoscale convective systems (organized clusters of thunderstorms) and tropical perturbations (tropical storms and hurricanes) which can account for a large portion of the yearly precipitation (Reyes et al. 1990). Further north, much of the interseasonal variability in precipitation in the Desert Southwest may be related to the gulf surge phenomenon described by Hales (1972). The gulf surge consists of a large mass of moist tropical air originating as far south as the mouth of the Gulf which moves rapidly northward (Figure 3). As the moist air mass crosses the elevated regions of the deserts of Sonora and Arizona, convective activity increases resulting in widespread precipitation events. Because there is no clear-cut relationship between large-scale circulation patterns over the region (however, see Stensrud et al. 1997), it is difficult to identify intra-seasonal variability of these surges; and, therefore, a possible important contributor to precipitation events in the Desert Southwest.
A number of researchers have examined changes in the synoptic-scale circulation and its effects on intraseasonal variability (Bryson and Lowry 1955; Carleton 1986, 1987). Through the examination of composited (averaged) synoptic-scale pressure patterns, it has become apparent that certain atmospheric configurations enhance (bursts) or stifle (breaks) convective activity in terms of cloudiness and precipitation (Carleton 1986, 1987). Carleton (1986, 1987) has also shown that widespread convective activity in the southwestern U.S. is typically associated with passing upper-level troughs (a pool of cooler air) in the Westerlies. In addition, the northward displacement of the subtropical ridge of high pressure and the formation of a "Four-Corners High" also results in increased thunderstorm activity in Arizona. A southerly displacement over northern Mexico of the subtropical ridge tends to lead to 'break' and drying out of the Desert Southwest. His conclusions suggest the importance of the location of the subtropical ridge in controlling convective activity in the southwestern U.S.
The scarceness of long-term precipitation records in the southwestern U.S. and northwestern Mexico has limited research on NA monsoon variability on historical time-scales (i.e., centuries). Nevertheless, marked variations on interseasonal and decadal time-scale for the NA monsoon have been noted (Carleton 1990; Green and Sellers 1964). Year-to-year variability in the NA monsoon activity may result from the occurrence of the same synoptic-scale patterns associated with interseasonal variability. Carleton (1990) has argued that the latitudinal shifts in the mid-level subtropical ridge over the southwest U.S., which account for a great deal of within-season varibility, are also responsible for year-to-year and decadal-scale variability. Northerly displacement of the subtropical ridge is associated with wetter Arizona summers, while a southerly shift coincides with decreases in summer total rainfall totals (Figure 4). The latitudinal shifting of the subtropical ridge is tied to phases of the Pacific-North American teleconnection pattern (PNA) (a strong trough and ridge pattern over the Northern Pacific and North America). Wetter summers are associated with the meridional phase (strong ridge and trough pattern) of PNA and its positive sea surface temperature anomalies which enhance ridging over North America. Drier summer have tended to follow zonal phases of the Pacific-North American teleconnection (weak ridge and trough pattern).
El Niño and Southern Oscillation(ENSO)-related Studies
ENSO-related phenomenon have been examined for possible causal mechanisms in NA monsoon variability. The relationships have proven to be ellusive for the summer rainfall totals in the monsoon region. Nevertheless, some studies have investigated the strength and geographic distribution of rainfall during and following ENSO events. Increased occurrence of tropical perturbations off the southwestern coast of Mexico has tended to follow the mature ENSO stage (northern hemisphere winter) when the intertropical convergence zone (ITCZ) is displayed farther northward and there is increased northward transport of water vapor (Reyes and Cadet 1988). In turn, northwestern Mexico, particularly for the Sonoran Desert region, has experienced positive anomalies in summer rainfall under these conditions (Reyes and Mejia-Trejo 1991). The greater occurrence of tropical perturbations and its relationship to the NA Monsoon phenomenon is still poorly understood. The ENSO relationship with summer precipitation in the southwestern United States is even weaker for this region. Andrade and Sellers (1988) found little correlation between ENSO and total summer rainfall in New Mexico and Arizona; however, Harrington et al. (1992) did not find a correlation between Southern Oscillation warm and cold events and the geographic distribution of summer precipitation in these same two states. Needless to say, further inquiries into ENSO-related phenomenon and the NA Monsoon are still required.
The NA monsoon and its associated summer precipitation vary on many different time and space scales. Typically, natural short-term climatic fluxuations are of the greatest importance in terms of increasing pressures on valuable resources such as water in this arid/semi-arid region. The scientific research outlined above has demonstrated the difficulties that lie within gauging the seasonal and interannual variability in monsoon precipitation. Causes of this shorter time variability appeared related to the frequency of Gulf Surges and the latitudinal position of the subtropical ridge. Research into the longer term variability (decadal to centuries) has been limited by lack of long-term instrumental records, particularly on the Mexican side. Though the records available have revealed no long-term temporal trends, decadal variability also appears to be related to longer term shifts in the subtropical ridge. Nevertheless, year-to-year fluxuations are typically the dominant component in the total variance of summer precipitation. The implications that human-induced climatic change may have for the Desert Southwest and northwestern Mexico remains a mystery. Given the variable nature of monsoon precipitation and the extreme dependency of the Desert Southwest on water availability, planners must be aware of the vagaries of the climatic variability of the NA Monsoon in order to prepare properly for them.
Adams, D.K. and A.C. Comrie. 1997, The North American Monsoon.
Bulletin of the American Meteorological Society, In press.
Andrade, E.R., Jr. and W. Sellers. 1988. El Niño and its effect on
precipitation in Arizona and western New Mexico. Journal and
Bryson, R.A., and W.P. Lowry. 1955. Synoptic climatology of the Arizona
summer precipitation singularity. Bulletin of the American
Meteorological Society 36:329-339.
Carleton, A.M. 1986. Synoptic-dynamic character of "bursts" and "breaks"
in the southwest U.S. summer precipitation singularity. Journal
of Climatology 6:605-623.
________1987. Summer circulation climate of the American Southwest: 1945-1984.
Annals of the Association of American Geographers 77:619-634.
________, D.A. Carpenter and P.J. Weber. 1990. Mechanisms of interannual
variability of the Southwest United States summer rainfall maximum.
Journal of Climate 3:99-1015.
Douglas, M.W., R. Maddox, K. Howard and S. Reyes. 1993. The Mexican Monsoon.
Journal of Climate 6:1665-1667.
Dunn, L. and J. Horel. 1994. Prediction of central Arizona convection.
Part I: Evaluation of the NGM and Eta model precipitation forecasts.
Weather and Forecasting 9:495-507.
Green, C. and W. Sellers. 1964. Arizona Climate. Tucson: University
of Arizona Press.
Hales, J. 1972. Surges of maritime tropical air northward over the Gulf
of California. Monthly Weather Review 100:298-306.
Harrington, J.A.,Jr., R. Cerveny, and R. Balling Jr. 1992. Impact of the
Southern Oscillation on the North American Southwest Monsoon.
Physical Geography 13:318-330.
Reyes, S. and L. Cadet. 1988. The southwest branch of the North
American Monsoon during 1979. Monthly Weather Review
_______and A. Mejia Trejo. 1991. Tropical perturbations in the eastern
Pacific and the precipitation field over northwestern Mexico in
relation to the ENSO phenomenon. International Journal of Climatology
Stensrud, D. 1997. Surges over the Gulf of California during the Mexican Monsoon. Monthly Weather Review, In Press.
Southwest Area Monsoon Project (SWAMP)
Pan American Climate Studies (PACS)