Abstracts of published papers
|Comrie, A.C., 1990: The climatology of
surface ozone in rural areas: a conceptual model. Progress
in Physical Geography 14, 295-316.
Ozone occurs both naturally and anthropogenically in the atmosphere, and several processes result in temporal and spatial variations in its concentration. Concern here is principally with surface ozone concentrations, although effects of ozone at other levels in the atmosphere are included where appropriate. The ozone problem will be approached via three phases of examination: firstly, the basic behavior of ozone, which comprises photochemical formation processes and relationships to meteorological variables; secondly, the sources of ozone, both background sources in the stratosphere and free troposphere, and anthropogenic sources in the planetary boundary layer (PBL) from urban and industrial plumes; thirdly, the transport of ozone within the PBL, not only the long range transport and accumulation of ozone in synoptic high pressure systems, but also subsynoptic local effects at the mesoscale. A conceptual model of the formation and transport of surface ozone in rural areas is formulated, and presented by way of a summary, with a brief discussion of promising research approaches and techniques. (from the introduction to the paper)
Comrie, A.C., 1992: A procedure for removing the synoptic climate signal from environmental data. International Journal of Climatology 12, 177-183.
Using weather-type frequencies from a synoptic climatology, a technique is presented that discriminates between within-type and between-type variations in a time series of climate-related environmental data. The removal of the synoptic climate signal, or declimatizing, is based on normalizing the data by the mean annual weather-type frequencies for the study period. Declimatizing is illustrated symbolically and with a worked hypothetical example. An application of the procedure to visibility data from Pittsburgh, Pennsylvania demonstrates its utility in decomposing complex climate-related environmental data into its component synoptic and non-synoptic influences. The methodology can also distinguish the relative importance of between-type and within-type changes in a synoptic climatology.
Comrie, A.C., 1992: An enhanced synoptic climatology of ozone using a sequencing technique. Physical Geography 13, 53-65.
A synoptic classification scheme is derived to examine basic associations between surface ozone pollution and the atmospheric circulation. Nine weather types are related to the daily maximum ozone concentration in Pittsburgh, Pennsylvania for the years 1978-1987. A sequencing technique is developed to extract the maximum utility from the classification scheme. An analysis of the sequences of synoptic weather types highlights additional spatial and temporal information, such as air mass origins, system speed and seasonal variations. Low concentrations of ozone are experienced in winter during lake-effect and cyclonic storms, which move in rapidly from the northwest bringing cold, cloudy, windy conditions with precipitation. High concentrations occur during summer in slow-moving anticyclones, with southwesterly transport and warm, sunny conditions that are favorable for photochemical formation of ozone. The study demonstrates that the use of a sequencing technique in conjunction with a synoptic classification scheme enables a more thorough analysis of the data.
Comrie, A.C. and Yarnal, B., 1992: Relationships between synoptic-scale atmospheric circulation and ozone concentrations in metropolitan Pittsburgh, Pennsylvania. Atmospheric Environment 26B, 301-312.
A synoptic climatology demonstrates the relationships between the atmospheric circulation and surface ozone (O3) concentrations. To deduce these associations, a subjective synoptic classification scheme is applied to ten years' O3 data from the Pittsburgh metropolitan area. The results focus on four aspects of the atmospheric circulation-O3 relationship: average, extreme-event, between-season, and year-to-year conditions. On average, each of the nine circulation types is related to a characteristic O3 concentration level and cumulative O3 dose. Extreme high-O3 events are associated with either the western side of a slowly-migrating anticyclone or a stagnating extended high-pressure ridge; low-O3 events are experienced under cool and cloudy cyclonic conditions. Between-season variations in the average and extreme circulation-O3 relationships are observed: the high-pressure features that produce the highest O3 levels in summer are related to low levels in winter, while circulation patterns that contribute very little to summertime O3 buildup are associated with the highest levels of wintertime O3. The latter situation could be caused by tropopause folding and the introduction of stratospheric ozone in winter months. While zonal (meridional) circulation regimes tend to produce lower (higher) mean annual O3 levels, such year-to-year changes in synoptic-type frequencies do not appear to be strongly related to interannual variations in O3, and other non-climatic factors appear to be of greater importance.
Comrie, A.C., 1994: A synoptic climatology of rural ozone pollution at three forest sites in Pennsylvania. Atmospheric Environment 28A, 1601-1614.
An analysis reveals strong relationships between ozone (O3) concentrations at three rural forest sites in north-central Pennsylvania and the synoptic-scale atmospheric circulation. To identify these associations, a synoptic classification scheme is applied to daily maximum 1h ambient surface O3 measurements for the growing seasons of 1988, 1989 and 1990. The results cover five aspects of the atmospheric circulationrural O3 relationship: overall conditions, O3 extremes, key weather sequences, the seasonal cycle and interannual differences. Overall, high rural O3 concentrations occur with southwesterly transport conditions on the western sides of anticyclones, while low values are found in post-frontal and cyclonic conditions. While slow-moving or stagnant anticyclones are occasionally associated with high-O3 episodes, these situations are most frequent in the same southwesterly transport regime. This behavior is the inverse of that found in Pittsburgh in a closely related study by Comrie and Yarnal (Atmospheric Environment Vol. 26B, No. 3, pp. 301-312, 1992). Unlike urban environments where air-mass stagnation leads to an episode, an episode in a non-urban environment requires transport of a polluted air mass from a source region. In this latter scenario, forest O3 levels are critically dependent on the air-mass history and trajectory. Key weather-pattern sequences show that the southwesterly transport must be preceded by stagnation of the air mass over an upwind polluted region, with stagnation and transport each lasting one to two days. The relative importance of these complementary mechanisms in the O3 climatology remains the same through the growing season. The unusually hot and dry conditions of the summer of 1988 were more favorable for O3 formation across all synoptic patterns, as compared to 1989 and 1990.
Comrie, A.C., 1994: Tracking ozone: air-mass trajectories and pollutant source regions influencing ozone in Pennsylvania forests. Annals of the Association of American Geographers 84 (4), 635-651.
Ground-level ozone pollution is causing measurable damage to the forests of the eastern United States, including those in Pennsylvania's Allegheny Plateau region. This area is surrounded by many urban and industrial pollution sources in the Midwest, Southeast, and Northeast United States and in southeast Canada. Any of these may play critical roles as source regions from which ozone and its precursor pollutants are carried toward the forests. This study identifies those geographic regions with the greatest potential influence on forest ozone concentrations via a climatological analysis of air-mass trajectories. In this analysis, observed meteorological data and a trajectory model are used to calculate the spatial history of polluted air-masses. Multiple trajectories are examined using a newly adapted methodology of ensemble trajectory analysis in combination with ozone data and key synoptic weather patterns from a related climatology. Results indicate a critical region of influence centered on the junction of the Ohio and Mississippi River Valleys, and extending eastward up the Ohio River Valley. These parts of Indiana, Ohio, Kentucky, Illinois and Missouri have the greatest likelihood of influencing high-ozone air masses arriving in Pennsylvania, and they coincide with some of the highest emission regions in the country. In the worst cases, air masses accumulate pollutants for several days as they stagnate over this region, and then continue accumulating pollutants as they move slowly toward Pennsylvania. Brief comments regarding the research and policy implications of these results are provided.
Comrie, A.C., 1996: An All-Season Synoptic Climatology of Air Pollution in the U.S.-Mexico Border Region.Professional Geographer 48 (3), 237-251.
The potential exists for widespread air quality problems in the U.S.-Mexico borderlands. Climate and weather are major factors governing the behavior of air pollution, and thus there is a need for greater understanding of border-region air pollution climatology. This paper presents a synoptic climatology of the 850 mb atmospheric circulation for the U.S.-Mexico border region, and an accompanying analysis of relationships between synoptic conditions and ground-level ozone. The synoptic methodology employs high-pass filtering to enable comparisons of all seasons, and it uses modified multiple k means clustering to identify six characteristic circulation patterns. The climatology succinctly summarizes important spatial and temporal complexities of border region circulation, including various pressure configurations, the seasonality of those patterns, and associated weather conditions across the region. These results are linked with ozone data for four border-region cities, and the subsequent findings highlight systematic seasonal and region-wide variations in ozone pollution corresponding to patterns of controlling climatic factors. Three high-ozone scenarios are identified, each of which selectively affects a different area or time of year.
Comrie, A.C., 1997: Comparing Neural Networks and Regression Models for Ozone Forecasting. Journal of the Air and Waste Management Association 47, 653-663.
Many large metropolitan areas experience elevated concentrations of ground-level ozone pollution during the summertime “smog season.” Local environmental or health agencies often need to make daily air pollution forecasts for public advisories and for inp ut into decisions regarding abatement measures and air quality management. Such forecasts are usually based on statistical relationships between weather conditions and ambient air pollution concentrations. Multivariate linear regression models have been w idely used for this purpose, and well-specified regressions can provide reasonable results. However, pollution-weather relationships are typically complex and nonlinear, especially for ozone -- properties that may be better captured by neural networks. This study investigates the potential for using neural networks to forecast ozone pollution, as compared to traditional regression models. Multiple regression models and neural networks are examined for a range of cities under different climate and ozone regimes, enabling a comparative study of the two approaches. Model comparison statistics indicate that neural network techniques are somewhat (but not dramatically) better than regression models for daily ozone prediction, and that all types of models are sensitive to different weather-ozone regimes and the role of persistence in aiding predictions.
Adams, D.K. and Comrie, A.C., 1997: The North American Monsoon. Bulletin of the American Meteorological Society, 78(10), 2197-2213.
The North American monsoon is an important feature of the atmospheric circulation over the continent, with a research literature that dates back almost one hundred years. We review the wide range of past and current research dealing with the meteorological and climatological aspects of the North American monsoon, highlighting historical development and major research themes. The domain of the North American monsoon is large, extending over much of the western United States from its region of greatest influence in northwestern Mexico. Regarding the debate over moisture source regions and water vapor advection into southwestern North America, there is general agreement that the bulk of monsoon moisture is advected at low-levels from the eastern tropical Pacific Ocean and the Gulf of California, while the Gulf of Mexico may contribute some upper-level moisture (although mixing occurs over the Sierra Madre Occidental). Surges of low-level moisture from the Gulf of California are a significant part of intra-seasonal monsoon variability, and they are associated with the configuration of upper-level mid-latitude troughs and tropical easterly waves at the synoptic scale, as well as the presence of low-level jets, a thermal low, and associated dynamics (including the important effects of local topography) at the mesoscale. Seasonally, the gulf surges and the latitudinal position of the mid-tropospheric subtropical ridge over southwestern North America appear to be responsible for much spatial and temporal variability in precipitation. Interannual variability of the North American monsoon system is high, but it is not strongly linked to El Niño or other common sources of interannual circulation variability. Recent mesoscale field measurements gathered during the South-West Area Monsoon Project (SWAMP) have highlighted the complex nature of the monsoon-related severe storm environment and associated difficulties in modeling and forecasting.
Comrie, A.C., 1998: Mapping the climatology of ozone potential for the U.S.-Mexico border region. Journal of the Arizona-Nevada Academy of Science 31(1), 1-12.
Concerns have arisen regarding the potential for ozone formation in the rapidly growing small and medium cities of the United States-Mexico border region. Most of these locations have limited or nonexistent ozone monitoring records, and yet for air quality planning purposes it would be very useful to know the susceptibility of such locations to urban ozone pollution. This paper presents estimates of susceptibility using two statistical measures, percentile rank and z-scores. Ozone potential is defined as the weather-related potential for ozone pollution, assuming a typically polluted urban atmosphere in the region. Ozone potential depends on the range of weather patterns that move over the area in question, and this study examines ozone potential using an existing synoptic climatology of six characteristic circulation patterns based on gridded 850 mb pressure-height data from 1963 to 1994. The synoptic catalog is augmented with matching 850 mb temperature data over the region. Ozone data for long periods of record (extending back to the mid-1970’s) are available for nine monitoring sites across the border region. Percentile rank and z-scores are used as relative measures of ozone concentration to determine ozone potential for each synoptic pattern, thus linking susceptibility to ozone pollution with the controlling atmospheric conditions. Maps of the results show that the border region is differentially susceptible to high-ozone weather conditions, leading to spatially and temporally distinct ozone patterns over the region. The spatial differences in susceptibility to urban ozone pollution are large, and are roughly equal to seasonal differences. Thus, the relative measures of ozone concentration applied in this study allow climate-related potentials for ozone pollution to be inferred for growing urban areas in the U.S.-Mexico border region that currently have sparse air quality data.
Comrie, A.C. and Glenn, E.C. 1998: Principal components-based regionalization of precipitation regimes across the Southwest United States and Northern Mexico, with an application to monsoon precipitation variability. Climate Research 10, 201-215.
We determine precipitation regions for the United States-Mexico border region based on seasonality and variability of monthly precipitation at 309 stations for the period 1961 to 1990. Using a correlation matrix of input data to avoid the effect of elevation on precipitation, we apply principal components analysis with oblique rotation to regionalize this large, climatologically complex study area. We examine the applicability of the method, two rules for defining region boundaries, the various defined regions themselves, and the effects of transforming input data and changing obliquity of component rotation. We obtain 9 consistent and largely contiguous regions from each of the analyses, including regions for the North American monsoon, the low deserts, the California Mediterranean region, and for summer precipitation regimes adjoining the Gulf of Mexico. The derived regions and associated boundaries make physical sense in terms of the driving atmospheric processes, and they are robust to transformed input data and changes in rotation procedures. The central border regions are remarkably consistent across analyses, with small changes to peripheral regions. We also identify 4 monsoon sub-regions, and we illustrate the applicability of the regionalization via an analysis of relationships between monsoon precipitation variability and 500 mb pressure heights. Significantly different 500 mb circulation patterns are associated with wet and dry monsoon seasons in each of the sub-regions, and it appears that shifts in 500 mb circulation relative to the geographic position of each sub-region influence seasonal precipitation variability, directly or indirectly. There are important differences between some sub-regions, but in general wet monsoons are associated with northward meridional bulging of the subtropical anticyclone over the continental monsoon areas, while dry monsoons are associated with zonal stretching of the subtropical anticyclone over adjacent oceans with slightly higher pressure-heights. Overall, the study provides a clear regionalization of the precipitation climatology for the southwest United States and northern Mexico, and shows its utility for studies of climate variability.
Comrie, A.C. and Diem, J.E. 1999: Climatology and forecast modeling of ambient carbon monoxide in Phoenix, AZ. Atmospheric Environment 33, 5023-5036.
We perform a climatology of factors influencing ambient carbon monoxide (CO), in which we examine the relationships between meteorology, traffic patterns, and CO at seasonal, weekly, and diurnal time scales in Phoenix, Arizona. From this analysis we identify a range of potentially important variables for statistical CO modeling. Using stepwise multivariate regression, we create a suite of models for hourly and 8-hour ambient CO designed for daily operational forecasting purposes. The resulting models include variables and interaction terms related to anticipated nocturnal atmospheric stability as well as antecedent and climatological CO behavior. The models are evaluated using a range of error statistics and skill measures. The most successful approach employs a two-stage modeling strategy in which an initial prediction is made that may, depending on the forecast value, be followed by a second prediction that improves upon the first. The best models provide accurate daily forecasts of CO, with explained variances approaching 0.9 and errors under 1 ppm.
Comrie, A.C. 2000: Mapping a wind-modified urban heat island in Tucson, Arizona (with comments on integrating research and undergraduate learning). Bulletin of the American Meteorological Society 81, 2417-2431.
Tucson, Arizona is an example of the many cities in the southwestern United States experiencing rapid growth and urban sprawl over the last several decades. The accompanying extensive modification of land use and land cover leads to many environmental impacts, including urban heat islands. The primary aim of this paper is to expand knowledge of the phenomenon for Tucson, by quantifying the amount of urban warming, and by mapping temperature patterns over the city and examining related aspects of the local scale atmospheric circulation. The secondary aim is to document how an applied empirical research project was integrated into an introductory undergraduate climatology class via active learning. The paper begins and concludes with general and practical comments on combining the research and educational aspects of the project.
An analysis of thirty-year temporal trends in urban and non-urban minimum temperatures across the region shows the rate of urban warming to be about three-quarters of the general regional warming. Tucson’s urban heat island is ~3°C over the last century, with >2°C of this warming in the last thirty years. The annual average urban warming trend over the last three decades is 0.071°C yr-1 with the strongest effect in March and the weakest effect in November. There is evidence that the latter is caused by strong, near-surface winds under stable conditions. A case study is presented comprising field measurements and map analysis of urban temperatures and supporting variables for February 13, 1999. Measurements include comprehensive mapping using vehicle-mounted thermistors and numerous local meteorological observations from around the city. Wind speeds during the field measurements were somewhat stronger than is typical of heat island studies, up to 12 m s-1. Nonetheless, because of terrain-induced flows and land surface heterogeneity, complex temperature patterns were observed. Several transient katabatic flows off surrounding mountain ranges were detected, leading to localized cold pockets. Locally warm areas in other parts of the city are associated with terrain sheltering or local land-surface heating. The central city showed a possible urban heating pattern with temperatures ~2°C higher than upwind rural air.
Diem, J.E. and Comrie, A.C. 2000: Integrating remote sensing and local vegetation information for a high resolution biogenic emissions inventory: application to an urbanized, semi-arid region. Journal of the Air and Waste Management Association 50, 1968-1979.
This paper presents a methodology for the development of a high resolution (30 m), standardized biogenic volatile organic compound (BVOC) emissions inventory and a subsequent application of the methodology to Tucson, Arizona. The region's heterogeneous vegetation cover cannot be modeled accurately with low resolution (e.g., 1 km) land cover and vegetation information. Instead, local vegetation data are used in conjunction with multi-spectral satellite data to generate a detailed vegetation-based land cover database of the region. A high resolution emissions inventory is assembled by associating the vegetation data with appropriate emissions factors. The inventory reveals a substantial variation in BVOC emissions across the region resulting from the region's diversity of both native and exotic vegetation. The importance of BVOC emissions from forest lands, desert lands, and the urban forest changes according to regional, metropolitan, and urban scales. Within the entire Tucson region, the average isoprene, monoterpenes, and OVOC fluxes are 454, 248, and 91 µg m -2 hr-1, respectively, with forest and desert lands emitting nearly all of the BVOCs. Within the metropolitan area, which does not include the forest lands, the average fluxes are 323, 181, and 70 µg m -2 hr-1, respectively. Within the urban area, the average fluxes are 801, 100, and 100 µg m-2 hr-1, respectively, with exotic trees such as eucalyptus, pine, and palm emitting most of the urban BVOCs. The methods presented in this paper can be modified to create detailed, standardized BVOC emissions inventories for other regions, especially those with spatially complex vegetation patterns.
Kolivras, K.N., Johnson, P.S., Comrie, A.C. and Yool, S.R. 2001: Environmental Variability and Coccidioidomycosis (Valley Fever). Aerobiologia 17, 31-42.
Coccidioidomycosis (valley fever) is a disease endemic to arid regions in the western hemisphere, and is caused by the soil-dwelling fungus Coccidioides immitis (C. immitis). In this paper, we provide an overview of the current state of knowledge regarding valley fever and C. immitis as related to climatic conditions and habitat requirements. Previous research shows there is a relationship between temperature and precipitation, and outbreaks of coccidioidomycosis. Incidence of the disease varies seasonally as well as annually due to changing climatic conditions. However, the specific environmental conditions that may produce an outbreak of coccidioidomycosis are not well understood in space and time. Previous studies have attempted to characterize C. immitis’ habitat. Temperature, moisture, salinity, and pH of the soil have all been considered separately in the geographic distribution of the fungus. Medical and proactive intervention are served best, however, by an integrative strategy that folds climate and surface variables into spatially-explicit models. We conclude with recommendations for future research directions.
E. Wright, A. Long, A. Comrie, S. Leavitt, T. Cavazos and C. Eastoe, 2001: North American monsoonal moisture sources revealed using temperature, precipitation, and precipitation stable isotope timeseries. Geophysical Research Letters 28, 787-790.
Results of analyses using timeseries of mean temperature, precipitation amount, and stable isotopes from precipitation from July-August in Tucson, Arizona, have revealed atmospheric circulation patterns related to the North American monsoon in the U.S. Southwest. The isotope timeseries and Tucson air temperatures and precipitation amount are significantly correlated. The temperature and isotope timeseries also correlate significantly with regional and extra-regional specific humidity, and with eastern Pacific SSTs near the Mexican coast, evidence for a dominantly Pacific/Gulf of California summer moisture source for the period 1983-1999. Separation of extra-regional wind vector datasets into groups of years matching relative isotopic depletion or enrichment of the Tucson July-August precipitation seasonal means for the stable isotope timeseries (usually the extreme years in the Tucson seasonal temperature means) suggest circulation patterns entraining more tropical moisture in cooler/isotopically depleted years, and entraining less tropical moisture in hotter/isotopically enriched years.
Diem, J.E. and Comrie, A.C. 2001: Air quality, climate, and policy: A case study of ozone pollution in Tucson, Arizona. The Professional Geographer, 469-491.
This article addresses the need to better understand the complex interactions between climate, human activities, vegetation responses, and surface ozone so that more informed air-quality policy recommendations can be made. The impacts of intraseasonal climate variations on ozone levels in Tucson, Arizona from April through September of 1995 to 1998 are determined by relating variations in ozone levels to variations in atmospheric conditions and emissions of ozone’s precursor chemicals, volatile organic compounds (VOCs) and nitrogen oxides (NOx), and by determining month-specific atmospheric conditions that are conducive to elevated ozone levels. Results show that the transport of ozone and its precursor chemicals within the Tucson area causes the highest ozone levels to be measured at a downwind monitor. The highest ozone levels occur in August, due in part to the presence of the North American monsoon. Atmospheric conditions conducive to elevated ozone concentrations differ substantially between the arid foresummer (May and June) and the core monsoon months ( July and August). Transport of pollution from Phoenix may have a substantial impact on elevated ozone concentrations during April, May, and June, while El Paso/Ciudad Juarez –derived pollution may contribute significantly to elevated ozone concentrations in August and September. Two broad policy implications derive from this work. Regional pollutant transport, both within the U.S. and between the U.S. and Mexico, is a potential issue that needs to be examined more intensively in future studies. In addition, spatiotemporal variations in sensitivities of ozone production require the adoption of both NOx and VOC control measures to reduce ozone levels in the Tucson area.
Yarnal, B., Comrie, A.C., Frakes, B. and Brown, D.P. 2001: Developments and prospects in synoptic climatology. International Journal of Climatology, 1923–1950.
Developments in synoptic climatology in the 1990s included advances in traditional synoptic climatology, empirical downscaling, and dynamical downscaling (i.e. regional climate modelling). The research emphasis in traditional, empirical–statistical approaches to synoptic climatology shifted from methodological development to applications of widely accepted classification techniques, including manual, correlation-based, eigenvector-based, compositing and indexing schemes. In contrast, most efforts in empirical downscaling, which became a well-established field of synoptic climatology during the 1990s, were directed to model development; applications were of secondary concern. Similarly, regional climate models (RCMs) burst onto the scene during the decade and focused on model development, although important progress was made in linking or coupling RCMs to regional or local surface climate systems. This paper discusses prospects for the future of traditional synoptic climatology, empirical downscaling and regional climate modelling. It concludes by looking at the present role of geographic information system (GIS) concepts in synoptic climatology and the potential future role of GIS to the field.
Diem, J.E. and Comrie, A.C. 2002: Allocating anthropogenic pollutant emissions over space: application to ozone pollution management. Journal of Environmental Management 63, 425-447.
An inventory of volatile organic compound (VOC) and nitrogen oxides (NOx) emissions is an important tool for the management of ground-level ozone pollution. This paper has two broad aims: it illustrates the potential of a geographic information system (GIS) for enhancing an existing spatially-aggregated, anthropogenic emissions inventory (EI) for Tucson, AZ, and it discusses the ozone-specific management implications of the resulting spatially-disaggregated EI. The main GIS-related methods include calculating emissions for specific features, spatially disaggregating region-wide emissions totals for area sources, and adding emissions from various point sources. In addition, temporal allocation factors enable the addition of a multi-temporal component to the inventory. The resulting inventory reveals that on-road motor vehicles account for approximately 50% of VOC and NOx emissions annually. On-road motor vehicles and residential wood combustion are the largest VOC sources in the summer and winter months, respectively. On-road motor vehicles are always the largest NOx sources. The most noticeable weekday vs. weekend VOC emissions differences are triggered by increased residential wood combustion and increased lawn and garden equipment use on weekends. Concerning the EI’s uncertainties and errors, on-road mobile, construction equipment, and lawn and garden equipment are identified as sources in the most need of further investigation. Overall, the EIs spatial component increases its utility as a management tool, which might involve visualization-driven analyses and air quality modeling.
Comrie, A.C. and Broyles, B., 2002: Variability and spatial modeling of fine-scale precipitation data for the Sonoran Desert of Southwest Arizona. Journal of Arid Environments 50, 573-592.
We present a unique new set of high spatial resolution precipitation data from a storage gauge network, for the sparsely observed northern Sonoran desert in southwest Arizona. We examine the nature and causes of the highly complex seasonal and spatial variability in the data, using fine-scale maps developed via spatial modeling and interpolation. These high-resolution maps had explained variances approaching 1.00, and precipitation errors of about 1 percent in winter and about 10 percent in summer. Seasonal precipitation ranges from near zero to almost 15 inches across the area, and shows high interannual variability. Localized convectional processes lead to summer anomalies that are more spatially complex than in winter when broad-scale synoptic and frontal processes cause precipitation. In general, summer and winter precipitation variability are tied to meridional-zonal shifts and east-west movement of the respective anticyclone or trough pattern over the region. Statistical links between major weather stations in the region and precipitation across the area are spatially inconsistent, especially in the west.
Diem, J.E. and Comrie, A.C. 2002: Predictive mapping of air pollution involving sparse spatial observations. Environmental Pollution 119, 99–117.
A limited number of sample points greatly reduces the availability of appropriate spatial interpolation methods.This is a common problem when one attempts to accurately predict air pollution levels across a metropolitan area. Using ground-level ozone concentrations in the Tucson,Arizona,region as an example, this paper discusses the above problem and its solution, which involves the use of linear regression. A large range of temporal variability is used to compensate for sparse spatial observations (i.e. few ozone monitors). Gridded estimates of emissions of ozone precursor chemicals, which are developed,stored,and manipulated within a geographic information system,are the core predictor variables in multiple linear regression models. Cross-validation of the pooled models reveals an overall R2 of 0.90 and approximately 7% error. Composite ozone maps predict that the highest ozone concentrations occur in a monitor-less area on the eastern edge of Tucson. The maps also reveal the need for ozone monitors in industrialized areas and in rural, forested areas.
Sheppard, P.R., Comrie, A.C., Packin, G.D., Angersbach, K., and Hughes, M.K. 2002: The climate of the US Southwest. Climate Research 21, 219-238.
This paper summarizes the current state of knowledge of the climate of the southwest USA (the 'Southwest'). Low annual precipitation, clear skies, and year-round warm climate over much of the Southwest are due in large part to a quasi-permanent subtropical high-pressure ridge over the region. However, the Southwest is located between the mid-latitude and subtropical atmospheric circulation regimes, and this positioning relative to shifts in these regimes is the fundamental reason for the region's climatic variability. Furthermore, the Southwest's complex topography and its geographical proximity to the Pacific Ocean, the Gulf of California, and the Gulf of Mexico also contribute to this region's high climatic variability. El Niño, which is an increase in sea surface temperature of the eastern equatorial Pacific Ocean with an associated shift of the active center of atmospheric convection from the western to the central equatorial Pacific, has a well developed teleconnection with the Southwest, usually resulting in wet winters. La Niña, the opposite oceanic case of El Niño usually results in dry winters for the Southwest. Another important oceanic influence on winter climate of the Southwest is a feature called the Pacific Decadal Oscillation (PDO), which has been defined as temporal variation in sea surface temperatures for most of the Northern Pacific Ocean. The combined effects of ENSO and PDO can amplify each other, resulting in increased annual variability in precipitation over the Southwest. The major feature that sets climate of the Southwest apart from the rest of the United States is the North American monsoon, which in the US is most noticeable in Arizona and New Mexico. Up to 50% of the annual rainfall of Arizona and New Mexico occurs as monsoonal storms from July through September. Instrumental measurement of temperature and precipitation in the Southwest dates back to the middle to late 1800s. From that record, average annual rainfall of Arizona is 322 mm [12.7 in.] while that of New Mexico is 340 mm [13.4 in.], and mean annual temperature of New Mexico is cooler (12 °C [53 °F]) than Arizona (17 °C [62 °F]). As instrumental meteorological records extend back only about 100–120 years throughout the Southwest, they are of limited utility for studying climate phenomena of long time frames. Hence, there is a need to extend the measured meteorological record further back in time using so-called "natural archive" paleoclimate records. Tree-ring data, which provide annual resolution, range throughout the Southwest, extend back in time for up to 1000 years or more in various forests of the Southwest, and integrate well the influences of both temperature and precipitation, are useful for this assessment of climate of the Southwest. Tree growth of mid elevation forests typically responds to moisture availability during the growing season, and a commonly used climate variable in paleo-precipitation studies is the Palmer Drought Severity Index (PDSI), which is a single variable derived from variation in precipitation and temperature. June–August PDSI strongly represents precipitation and, to a lesser extent, temperature of the year prior to the growing season (prior September through current August). The maximum intra-ring density of higher elevation trees can yield a useful record of summer temperature variation. The combined paleo-modern climate record has at least three occurrences of multi-decadal variation (50–80 years) of alternating dry (below average PDSI) to wet (above average PDSI). The amplitude of this variation has increased since the 1700s. The most obvious feature of the temperature record is its current increase to an extent unprecedented in the last four hundred years. Because this warming trend is outside the variation of the natural archives, it is possible that anthropogenic impacts, such as increased atmospheric concentrations of greenhouse trace gases, are playing a role in climate of the Southwest. Accordingly, this pattern merits further research in search of its cause or combination of causes.
Cavazos, T., Comrie, A.C. and Liverman, D.M., 2002: Intraseasonal variability associated with wet monsoons in southeast Arizona. Journal of Climate 15, 2477-2490.
The intraseasonal evolution of the North American monsoon in southeast Arizona during the 1980-1993 period is investigated using a neural network-based nonlinear classification technique known as the self-organizing map (SOM). The goal of the SOM algorithm is to discover meaningful low-dimensional structures hidden in the high-dimensional observations. Various daily lagged atmospheric fields (850-hPa meridional winds, 700-hPa specific humidity, 500-hPa geopotential heights, and 850-500-hPa thickness) for the summer season (Jun-Jul-Aug-Sep) of the 1980-1993 period are used in the nonlinear classification of monsoon modes. Special emphasis is given to the wettest monsoon modes. The neural network classification successfully captures the multidimensional interaction of the atmospheric variables during the monsoon evolution, and shows monsoon “bursts” and “breaks” in a given year. Spectral analysis of daily summer rainfall in the study area reveals a significant peak in the 12-18-day band; a secondary and significant peak is also found near 40 days. Thus, monsoon bursts and breaks seem to be modulated by low frequency variability.
The SOM nonlinear classification shows that the mature phase of the monsoon is associated with two distinct intraseasonal (>10 days) wet monsoon modes. The signature of the wettest monsoon mode is a zonal three-cell anomalous mid-tropospheric height pattern over the North Pacific-North American sector, suggesting a large-scale dynamical mechanism, possibly linked to sea surface temperature (SST) anomalies in the North Pacific. This zonal mode, which is most frequent in July and August, is characterized by an enhanced and northeastward-displaced monsoon ridge, large amounts of mid-tropospheric moisture over the study area, and an out of phase relationship between precipitation in the Southwest United States and precipitation in the Great Plains. The zonal mode has been recognized in longer data sets and it is the most typical mode that characterizes the mature phase of the monsoon in the Southwest United States. In contrast, the second wettest intraseasonal monsoon mode does not show a monsoon ridge, but a meridional three-cell anomalous mid-tropospheric height pattern along the West coast of North America, weak height anomalies over the rest of North America, and large amounts of moisture over the study area. Importantly, this meridional mode, which is most frequent in August and September, does not show out of phase links to Great Plains precipitation. The meridional wet mode also shows an anomalous low-level cyclonic circulation off the west coast of central-south Mexico suggesting that convective activity off the southern Mexican coast – possibly associated with the intertropical convergence zone – may cross over the Isthmus of Tehuantepec toward the Gulf of Mexico and the southern United States. This would explain the weak link between precipitation in the Southwest and precipitation in the Great Plains during August and September of the 1980-1993 period.
At more regional scales, the zonal wet mode is also characterized by a latitudinal gradient of SST anomalies between Baja California and southern Mexico and reversed low-level flow over the Gulf of California. Looking at extreme wet monsoons outside of the study period (e.g., 1955, 1959, 1999) indicate that the positive SST anomaly pattern along the Pacific coast of Baja California, which characterized wet events during 1980-1993, can be completely reversed during other extreme wet events. These contrasting results suggest that interaction between local and remote forcing mechanisms over the study area are complex during extreme events and need further investigation.
Brown DP and AC Comrie (2002) Spatial modeling of winter temperature and precipitation in Arizona and New Mexico, U.S.A. Climate Research 22, 115-128.
The development of a statistical modeling technique suitable for producing mean and interannual gridded climate datasets for a topographically varying domain is undertaken. Stepwise regression models at 1x1 km resolution are generated to estimate mean winter temperature and precipitation for the Southwest United States for the years 1961 to 1990. Topographic predictor variables are used to explain spatial variance in the datasets. Kriging and inverse distance weighting interpolation algorithms are utilized to account for model residuals. The final regression models show a high degree of explained variance for temperature (R2 = 0.98, MBE = -0.15° C, RMSE = 0.74° C) and a moderate degree of explained variance for precipitation (R2 = 0.63, MBE = -1.4 mm, RMSE = 27.0 mm). Several smaller-scale precipitation regression models are developed for comparison to the domain-wide model, but do not show marked accuracy improvements. Observed values of winter temperature and precipitation from the years 1961 to 1999 are compared to the 30-year modeled means, and the differences are interpolated using kriging (temperature) and inverse distance weighting (precipitation). The result is a 39-year time series of maps and datasets of winter temperature and precipitation at 1x1 km resolution for the Southwest United States.
Ni, F., Cavazos, T., Hughes, M.K., Comrie, A.C. and Funkhouser, G., 2002: Cool season precipitation in the Southwestern United States since AD 1000: Comparison of linear and nonlinear techniques for reconstruction. International Journal of Climatology, 22, 1645-1662.
A 1000-year reconstruction of cool-season (November-April) precipitation was developed for each climate division in Arizona and New Mexico from a network of 19 tree-ring chronologies in the Southwestern United States. Linear regression (LR) and artificial neural networks (NN) models were used to compare the response of tree growth to cool-season precipitation. The stepwise LR model was cross-validated with a leave-one-out procedure while the NN was validated with a bootstrap technique using 1931-1988 records. The final models were also independently validated using the 1896-1930 precipitation data. In most of the climate divisions both techniques can successfuly simulate dry and normal years, and the NN seems to better capture large precipitation events and more variability than the LR. In the 1000-year reconstructions the NN also produces more distinctive wet events and more variability, while the LR produces more distinctive dry events. The 1000-year reconstructed precipitation from the two models shows several sustained dry and wet periods comparable to the 1950s drought (e.g., 16th century megadrought) and the post-1976 wet periods (e.g., 1330s, 1610s). The impact of extreme periods on the environment may be stronger during sudden reversals from dry to wet, which were not uncommon throughout the millennium, such as the 1610s wet interval that followed the 16th century megadrought. The instrumental records suggest that strong dry to wet precipitation reversals in the past 1000 years might be linked to strong shifts from cold to warm El Nino/Southern Oscillation (ENSO) events and from negative to positive Pacific Decadal Oscillation (PDO).
Brown, D.P. and Comrie, A.C., 2002: Sub-regional seasonal precipitation linkages to SOI and PDO in the Southwest United States. Atmospheric Science Letters, 3, 94-102.
This paper highlights the relationship between precipitation variability at the sub-regional level in the Southwest United States and the SOI and PDO climate teleconnection indices during the period 1950 – 2000. Statistical correlations at the a = 0.05 and a = 0.01 levels are calculated for fall, winter, and spring precipitation in the Southwest, and contemporaneous and antecedent seasonal SOI and PDO index values. A strong SOI-winter precipitation signal is seen to progress across Arizona and New Mexico from southwest to northeast over a three-season lagged period. The PDO also exhibits a strong relationship with winter and spring precipitation in New Mexico; however, the PDO is not well correlated with precipitation in Arizona. The results underscore the non-uniform spatio-temporal relationships of the SOI and PDO indices as they relate to the precipitation regime of the Southwest, and provide a framework for future diagnostic analyses of these relationships.
Komatsu, K., Vaz, V., McRill, C., Colman, T., Comrie, A., Sigel, K., Clark, T., Phelan, M., Hajjeh, R. and Park, B., 2003: Increase in coccidioidomycosis –
(Summary from introduction – no abstract) Coccidioidomycosis is a systemic infection caused by inhalation of airborne spores from Coccidioides immitis, a fungus found in soil in the southwestern
Kolivras, K.N. and Comrie, A.C., 2003: Modeling valley fever (coccidioidomycosis) incidence based on climate conditions. International Journal of Biometeorology 47, 87-101.
Valley fever (coccidioidomycosis) is a disease endemic to arid regions within the western hemisphere, and is caused by a soil-dwelling fungus, Coccidioides immitis. Incidence data for Pima County reported to the Arizona Department of Health Services as new cases of valley fever were used to conduct exploratory analyses and develop monthly multivariate models of relationships between valley fever incidence and climate conditions and variability in Pima County, Arizona, U.S.A. Bivariate and compositing analyses conducted during the exploratory portion of the study revealed that antecedent temperature and precipitation in different seasons are important predictors of incidence. These results were used in the selection of candidate variables for multivariate predictive modeling, which was designed to predict deviation from mean incidence based on past, current, and forecast climate conditions. The models were specified using a backward stepwise procedure, and were most sensitive to key predictor variables in the winter season and variables that were time-lagged one year or more prior to the month being predicted. Model accuracy was generally moderate (R2 values for the monthly models tested on independent data ranged from 0.15 to 0.50), and months with high incidence can be predicted more accurately than months with low incidence.
Kolivras, K.N. and Comrie, A.C., 2004: Climate and infectious disease in the southwestern
As in many parts of the world, climate variability has a strong impact on infectious diseases within the southwestern USA. Moisture and temperature conditions can either indirectly impact disease by providing an environment conducive to the growth of an animal host or reservoir, or directly through the survival and dispersal of an infectious agent. It is also expected that climate change will affect the number of cases and/or the spatial distribution of infectious diseases. Before the effects of climate change on diseases can be determined, an understanding of the basic relationship between incidence and climate variability should be established. A review of climate impacts on four infectious diseases (hantavirus, plague, dengue and coccidioidomycosis) currently found in southwestern USA (or potentially found in the southwest in the case of dengue) is followed by suggested future research to further understand the relationship between climate variability/change and disease.
Abraham, J.S. and Comrie, A.C., 2004: Real-time ozone mapping using a regression-interpolation hybrid approach, applied to
Real-time ozone (O3) maps, intended for public access and mass media, are generated from spatially interpolating (i.e., kriging) sparse monitoring data and are typically characterized by over-smoothed surfaces that inadequately represent local-scale spatial patterns (e.g., averaged over 1 km2). In this paper, a hybrid regressioninterpolation methodology is developed to enhance the representation of local-scale spatiotemporal patterns with an application to Tucson, Arizona. The mapping of local patterns is enhanced with pre-interpolation regression modeling of local-scale deviation-from-mean variability, preserving variation in the monitor data that is ubiquitous across the modeling domain (i.e., the areal mean). The model is trained on several years of deviation-frommean hourly O3 data, and predictor variables are developed using theoretically and empirically derived proxy regression variables. The regression model explains a significant proportion of the variation in the data (r2 = 0.54), with an average error of 7.1 ppb. When augmented with the areal mean, the r2 of the pre-interpolation model increases to 0.847. Model residuals are then spatially interpolated to the extents of the modeling domain. Final concentration estimate maps are the summation of areal mean, regression, and spatially interpolated surfaces, preserving absolute values at monitor locations.
Brown, D.B. and Comrie, A.C., 2004: A winter precipitation ‘dipole’ in the Western United States associated with multidecadal ENSO variability. Geophysical Research Letters 31, doi:10.1029/2003GL018726.
The variability of winter precipitation across the western United States has important implications for a wide range of physical and socioeconomic systems. While El Nino-Southern Oscillation (ENSO) teleconnections explain a high degree of interannual variance in western U.S. winter precipitation, their influence on decadal time scales is less well understood. In this study, we examine the relationship between ENSO conditions and winter precipitation in the western U.S. within the context of decadal-scale variability, as represented by phasing of the Pacific Decadal Oscillation (PDO). We identify spatial inconsistencies in the ENSO-precipitation relationship, commensurate with PDO phase shifts, which take the form of a ‘dipole’ signature across the western U.S. This finding has implications for the knowledge of uncertainty of ENSO teleconnections, and may prove meaningful for users of climate information throughout the region.
Kliman, S.S. and Comrie, A.C., 2004: Effects of vegetation on residential energy consumption. Home Energy, July/August, 38-42.
This paper does not have an abstract. A summary is provided here: We conducted an empirical study of 105 existing homes in the Metropolitan Tucson area. The study examined and quantified the actual relationship between vegetation and residential energy consumption in a hot dry environment. The study homes were a mix of masonry (high-mass) construction, generally built between 1930 and the late 1970s, and frame and stucco (low-mass) construction, generally built in the 1980s and 1990s. Data were collected from a variety of sources in an effort to obtain as much information as possible about the study homes. Homeowner surveys collected information about the physical structure, such as construction type, age, size, and color of the house, type of heating and cooling equipment, any amenities which would impact energy consumption (such as pools and spas), and the type of thermostat (programmable versus non-programmable). Homeowners were asked to document the number of hours the house is occupied during a typical week and weekend. They were also asked to document their typical daytime and nighttime thermostat settings for both summer and winter. This information included whether and how they adjust the thermostat or mechanical equipment when the home is not occupied. The final section of the survey pertained to the landscaping. Homeowners were asked to complete a matrix of typical landscape materials and the four cardinal directions to document the landscape around their home. They were also asked to provide a simple sketch of the home and adjacent landscape, including the location of the front door and the orientation of the house. The physical characteristics of the house, such as wall construction, exterior color, roof type, and the type of cooling equipment, combined with the living habits of the occupants, in particular how they set their thermostats, far outweighed the impacts of vegetation. While the computer simulation studies predict ideal average summer energy savings of 7%-8% from the planting of trees, in our actual homes, the other real world effects, such as thermostat settings, obscured any measurable effect from the vegetation. The analysis of houses included in this study—existing homes with typical landscaping patterns—was unable to document any measurable savings from vegetation, whether trees, shrubs, grass, or natural desert. The fact that none of the vegetation variables employed, whether shade trees or well-watered grass, provided a quantifiable savings indicates that neither shading benefits nor evapotranspiration benefits were realized. The results did, however, confirm the negative impact from trees on the winter heating load documented in previous studies.
Crimmins, M. A. and Comrie, A.C., 2004: Interactions between antecedent climate and wildfire variability across southeast Arizona. International Journal of Wildland Fire 13, 455-466.
Long-term antecedent climate conditions are often overlooked as important drivers of wildfire variability. Fuel moisture levels and fine-fuel productivity are controlled by variability in precipitation and temperature at long timescales (months to years) before wildfire events. This study examines relationships between wildfire statistics (total area burned and total number of fires) aggregated for south-eastern Arizona and antecedent climate conditions relative to 29 fire seasons (April–May–June) between 1973 and 2001. High and low elevation fires were examined separately to determine the influence of climate variability on dominant fuel types (low elevation grasslands with fine fuels v. high elevation forests with heavy fuels). Positive correlations between lagged precipitation and total area burned highlight the importance of climate in regulating fine fuel production for both high and low elevation fires. Surprisingly, no significant negative correlations between precipitation and seasonal wildfire statistics were found at any seasonal lag. Drought conditions were not associated with higher area burned or a greater number of fires. Larger low elevation fires were actually associated with wet antecedent conditions until just before the fire season. Larger high elevation fires were associated with wet conditions during seasons up to 3 years before the fire season.
Wise, E.K. and Comrie, A.C., 2005: Meteorologically-adjusted urban air quality trends in the southwestern United States. Atmospheric Environment 39, 2969-2980.
Cities in the Southwestern United States (Southwest) are often close to violating tropospheric ozone (ozone) and particulate matter (PM) federal air quality standards, and local climate and weather conditions play a large part in determining whether or not pollutant levels exceed the federally mandated limits and by what magnitude. The Kolmogorov–Zurbenko (KZ) filter method has been used in a number of studies in the Eastern United States to determine meteorological controls on ozone concentrations and to separate out those effects in order to examine underlying trends. The Southwest, however, experiences a different climate regime than other parts of the country, and atmospheric controls on air quality in the region have not been examined in this manner. This paper determines which meteorological variables most influence ozone and PM in the Southwest and examines patterns of underlying pollutant trends due to emissions. Ozone and PM data were analyzed over the time period 1990–2003 for the Southwest’s five major metropolitan areas: Albuquerque, NM; El Paso, TX; Las Vegas, NV; Phoenix, AZ; and Tucson, AZ. Results indicate that temperature and mixing height most strongly influence ozone conditions, while moisture levels (particularly relative humidity) are the strongest predictors of PM concentrations in all five cities examined. Meteorological variability typically accounts for 40–70% of ozone variability and 20–50% of PM variability. Long-term ozone trends are highly variable, but records from most stations indicate increasing concentrations over the last decade. Long-term trends in PM concentrations were relatively flat in all five cities analyzed but contained coincident extremes unrelated to meteorology.
Park, B.J., Sigel, K., Vaz, V., Komatsu, K., McRill, C., Phelan, M., Colman, T., Comrie, A.C., Warnock, D.W., Galgiani, J.N. and Hajjeh, R.A., 2005: An epidemic of coccidioidomycosis in Arizona associated with climate changes, 1998-2001. Journal of Infectious Diseases, 191, 1981-1987.
Background: Coccidioidomycosis case-reports in Arizona have increased substantially. We investigated factors associated with the increase. Methods: We analyzed the National Electronic Telecommunications System for Surveillance (NETSS) from 1998 to 2001 and used Geographic Information Systems (GIS) to map high incidence areas in Maricopa County. Poisson regression analysis was performed to assess the effect of climatic and environmental factors on monthly cases; a model was developed and tested to predict outbreaks. Results: Overall incidence in 2001 was 43 cases/ 100,000 population, a significant (p<0.01 for trend) increase from 1998 (33/ 100,000); the highest age-specific rate was in persons >65 years old (79/ 100,000 in 2001). Analysis of NETSS data by season indicated high incidence periods during the winter (November- February). GIS showed the highest incidence areas were in the periphery of Phoenix. Multivariable Poisson regression modeling revealed a combination of certain climatic and environmental factors were highly correlated with seasonal outbreaks (R2= 0.75). Conclusions: Coccidioidomycosis in Arizona has increased. Its incidence is driven by seasonal outbreaks associated with environmental and climatic changes. Our study may allow public health officials to predict seasonal outbreaks in Arizona, and alert the public and physicians early to implement appropriate preventive measures.
Wise, E.K. and Comrie A.C., 2005: Extending the KZ filter: application to ozone, particulate matter and meteorological trends. Journal of the Air and Waste Management Association 55, 1208-1216.
Tropospheric ozone (ozone) and particulate matter (PM) are pollutants of great concern to air quality managers. Federal standards for these pollutants have been promulgated in recent years due to the pollutants’ known adverse effects on human health, the environment, and visibility. Local meteorological conditions exert a strong influence over day-to-day variations in pollutant concentrations; therefore, the meteorological signal must be removed in order for air quality planners and managers to examine underlying emissions-related trends and make better air quality management decisions for the future. Although the Kolmogorov-Zurbenko (KZ) filter has been widely used for this type of trend separation in ozone studies in the eastern United States, this paper aims to extend the method in three key ways. First, while the KZ filter is known as a useful tool for ozone analysis, this study also evaluates its effectiveness when applied to PM. Second, the method was applied to Tucson, Arizona, a city in the semi-arid southwestern United States (Southwest), in order to evaluate the appropriateness of the method in a region with weaker synoptic weather controls on air quality than the eastern United States. Third, additional forms of output were developed and tailored to be more applicable to decision-makers’ needs through a partnership between academic researchers and air quality planners and managers. Results of the study indicate that the KZ filter is a useful method for examining emissions-related PM trends, resulting in small, but potentially significant, differences after adjustment. For the Tucson situation with weaker synoptic controls, the KZ method identified mixing height as a more important variable than has been found in other cities.
Comrie, A.C., 2005: Climate factors influencing coccidioidomycosis seasonality and outbreaks. Environmental Health Perspectives 113, 688-692.
Although broad links between climatic factors and coccidioidomycosis have been established, the identification of simple and robust relationships linking climatic controls to seasonal timing and outbreaks of the disease have been elusive and remain poorly understood. Using an adaptive data-oriented method for estimating date of exposure, this paper analyzes hypotheses linking climate and dust to fungal growth and dispersion and evaluates their respective roles for Pima County, Arizona. Results confirm a strong bimodal disease seasonality that was suspected but not previously seen in reported data. Dispersion-related conditions are important predictors of coccidioidomycosis incidence during fall, winter and the arid foresummer. However, precipitation during the normally arid foresummer 1.5-2 years prior to the season of exposure is the dominant predictor of the disease in all seasons, accounting for half of the overall variance. Cross-validated models combining antecedent and concurrent conditions explain 80 percent of the variance in coccidioidomycosis incidence.
Ray, A.J., Garfin, G.M., Wilder, M., Vásquez-León, M., Lenart, M. & Comrie, A.C., 2007: Applications of monsoon research: Opportunities to inform decisionmaking and reduce regional vulnerability. Journal of Climate 20, 1608–1627.
This article presents current efforts to understand the interactions between the North American Monsoon and society, in order to develop applications for monsoon research in a diverse, multicultural, and binational region. The North American Monsoon is an annual precipitation regime beginning in early June in Mexico and progressing northward to the southwestern United States. The region includes stakeholders in large urban complexes, productive agricultural areas, and vast areas of relatively open arid to semi-arid ecosystems. The U.S.-Mexico border and cultural and socio-economic factors result in a patchwork of sensitivities and capacities to cope with variability in the physical system and to forecasts thereof. We review methodologies to link climate science with society and to study vulnerability in the monsoon region. The article highlights four principal sectors in which there is an opportunity for monsoon science to benefit society including: natural hazards management, agriculture, public health, and water management. We synthesize a list of common stakeholder needs and a calendar of decisions for which timely information is needed. We argue that there is a need to deliberately link monsoon research to integrated assessments involving both scientists and stakeholders in the region. Mechanisms should be established to ensure and coordinate 1) sector-specific assessments research, 2) user-centric experiments to develop useable products and respond to stakeholder feedback, 3) outreach and extension. We expect that coordinated applications and assessments efforts can capitalize on the opportunities for monsoon science to inform decisionmaking, and, in the best instances, reduce regional climate-related vulnerabilities and enhance regional sustainability.
Comrie, A.C. 2007: Climate Change and Human Health. Geography Compass 1, 325–339.
What kinds of climate-mediated diseases exist, and how are projected climate changes expected to alter their spread and timing? Disease is produced in a complex way, through coupled interactions between natural and human systems. Climate is a major factor controlling ecosystem variability and therefore the potential for outbreaks of certain diseases. Yet the concept of vulnerability shows how overall disease risk depends not only on the environmental exposure, but also on the sensitivity and adaptive capacity of the group and place experiencing it. These interactions between environment and society are highlighted through a set of climate-related diseases, ranging from direct to complex relationships, including extreme heat, air pollution, aeroallergens, fungi, water- and food-borne diseases, influenza, rodent-borne diseases, and insect-borne diseases.
Tamerius, J., Wise, E.K., Uejio, C.K., McCoy, A., and Comrie, A.C. 2007: Climate and human health: synthesizing environmental complexity and uncertainty. Stochastic Environmental Research and Risk Assessment (SERRA) 21, 601-613.
Broad relationships between weather and human health have long been recognized, and there is currently a large body of research examining the impacts of climate change on human health. Much of the literature in this area examines climate–health relationships at global or regional levels, incorporating mostly generalized responses of pathogens and vectors to broad changes in climate. Far less research has been done to understand the direct and indirect climate-mediated processes involved at finer scales. Thus, some studies simplify the role of climate and may over- or underestimate the potential response, while others have begun to highlight the subtle and complex role for climate that is contingent on other relevant processes occurring in natural and social environments. These fundamental processes need to be understood to determine the effects of past, current and future climate variation and change on human health. We summarize the principal climate variables and climate-dependent processes that are believed to impact human health across a representative set of diseases, along with key uncertainties in these relationships.
Kolivras, K.N. and Comrie, A.C., 2007: Regionalization and variability of precipitation in Hawaii. Physical Geography 21, 76-96.
Regions based on seasonal precipitation variability for Hawaii are determined using a principal components analysis applied to 124 stations for the period 1971-2000. Nine regions are delineated and are consistent with known precipitation patterns; leeward and windward stations are in separate regions on all islands. Within each region, the relationship between precipitation and the El Niño-Southern Oscillation (ENSO) is examined using a correlation analysis with the Southern Oscillation Index (SOI), and the Niño 3.4 and Niño 1+2 indices. Precipitation is most frequently correlated with ENSO in the different regions using SOI and Niño 3.4. Using several nonparametric statistical tests, it is determined that while average precipitation received in Hawaii during El Niño events is significantly different from average precipitation (1971-2000) and from precipitation received during La Niña events, the relationship between precipitation and individual ENSO events within regions is rarely significant. Finally, during El Niño or La Niña events, average precipitation receipt across the regions co-varies during winter and summer under concurrent conditions and a one-season lag. Synoptic patterns are examined and indicate a deviation from average conditions during ENSO events that affects subsidence and precipitation patterns.
Comrie, A.C. and Glueck, M.F. 2007: Model Sensitivity for Assessing Climatologic Effects on the Risk of Acquiring Coccidioidomycosis. Annals of the New York Academy of Sciences 1111, 83–95.
Understanding the predictive relationships between climate variability and coccidioidomycosis is of great importance for the development of an effective public health decision-support system. Preliminary regression-based climate modeling studies have shown that about 80% of the variance in seasonal coccidioidomycosis incidence for southern Arizona can be explained by precipitation and dust-related climate scenarios prior to and concurrent with outbreaks. In earlier studies, precipitation during the normally arid foresummer 1.5–2 years prior to the season of exposure was found to be the dominant predictor. Here, the sensitivity of the seasonal modeling approach is examined as it relates to data quality control (QC), data trends, and exposure adjustment methodologies. Sensitivity analysis is based on both the original period of record, 1992–2003, and updated coccidioidomycosis incidence and climate data extending the period of record through 2005. Results indicate that models using case-level data exposure adjustment do not suffer significantly if individual case report data are used “as is.” Results also show that the overall increasing trend in incidence is beyond explanation through climate variability alone. However, results also confirm that climate accounts for much of the coccidioidomycosis incidence variability about the trend from 1992 to 2005. These strongly significant relationships between climate conditions and coccidioidomycosis incidence obtained through regression modeling further support the dual “grow and blow” hypothesis for climate-related coccidioidomycosis incidence risk
Bieda III, S.W., Castro, C.L., Mullen, S.L., Comrie, A.C. and Pytlak, E. 2009: The relationship of transient upper-level troughs to variability of the North American monsoon system. Journal of Climate 22, 4213-4227.
Relationships between transient upper-tropospheric troughs and warm season convective activity over the southwest United States and northern Mexico are explored. Analysis of geopotential height and vorticity fields from the North American Regional Reanalysis and cloud-to-ground lightning data indicates that the passage of mobile inverted troughs (IVs) significantly enhances convection when it coincides with the peak diurnal cycle (1800–0900 UTC) over the North American monsoon (NAM) region. The preferred tracks of IVs during early summer are related to the dominant modes of Pacific sea surface temperature (SST) variability. When La Niña–like (El Niño–like) conditions prevail in the tropical Pacific and the eastern North Pacific has a horseshoe-shaped negative (positive) SST anomaly, IVs preferentially track farther north (south) and are slightly (typically one IV) more (less) numerous. These results point to the important role that synoptic-scale disturbances play in modulating the diurnal cycle of precipitation over the NAM region and the significant impact that the statistically supported low-frequency Pacific SST anomalies exert on the occurrence and track of these synoptic transients.
Comrie, A.C. 2010: Nietzsche's challenge to physical geography. ACME 9 (1), 34-46.
Using the philosophy of Nietzsche as a stimulus, I aim to engage physical geographers and fellow scientists to reconsider their roles as scientists and to make their work more action-oriented and powerful. I outline the false mystique of science and the misconception of seeing science as independent of people and society. I make a case that science gains its power by the way we attach meaning to it and its findings, and that we should act on our ability to bestow that power. Through Nietzsche, I argue that we are challenged to overcome our trained tendency toward detached environmental science and instead put in place a new physical geography that includes meaning and action. We have the opportunity to do so in practical ways, by being reflexive and acknowledging the context of our science, and by finding more ways to communicate our ideas in support of action to change our world.
Morin, C. and Comrie, A.C. 2010: Modeled response of the West Nile virus vector Culex quinquefasciatus to changing climate using the dynamic mosquito simulation model. International Journal of Biometeorology 54,517-529, doi:10.1007/s00484-010-0349-6.
Climate can strongly influence the population dynamics of disease vectors and consequently it is a key component of disease ecology. Future climate change and variability may alter the location and seasonality of many disease vectors, possibly increasing the risk of disease transmission to humans. The mosquito species Culex quinquefasciatus is a concern across the southern United States because of its role as a West Nile virus vector and its affinity for urban environments. Using established relationships between atmospheric variables (temperature and precipitation) and mosquito development, we have created the Dynamic Mosquito Simulation Model (DyMSiM) to simulate Cx. quinquefasciatus population dynamics. The model is driven with climate data and validated against mosquito count data from Pasco County, Florida and Coachella Valley, California. Using one-week and two-week filters, mosquito trap data are reproduced well by the model (p< 0.0001). Dry environments in southern California produce different mosquito population trends than moist locations in Florida. Florida and California mosquito populations are generally temperature-limited in winter. In California, locations are water-limited through much of the year. Using future climate projection data generated by the National Center for Atmospheric Research CCSM3 general circulation model, we applied temperature and precipitation offsets to the climate data at each location to evaluate mosquito population sensitivity to possible future climate conditions. We found that temperature and precipitation shifts act interdependently to cause remarkable changes in modeled mosquito population dynamics. Impacts include a summer population decline from drying in California due to loss of immature mosquito habitats, and in Florida a decrease in late-season mosquito populations due to drier late summer conditions.
Delgado, S., Neyra, R.C., Machaca, V.Q., Juarez, J.A., Chu, L.C., Verastegui, M., Apaza, G.M., Bocángel, C.D., Tustin, A.W., Sterling, C.R., Comrie, A.C., Naquira, C., del Carpio, J.C., Gilman, R., Bern, C. and Levy, M. 2011: A history of Chagas disease transmission, control, and re-emergence in peri-rural La Joya, Peru. PLoS Neglected Tropical Diseases, 5(2): e970. doi:10.1371/journal.pntd.0000970.
Background: The history of Chagas disease control in Peru and many other nations is marked by scattered and poorly documented vector control campaigns. The complexities of human migration and sporadic control campaigns complicate evaluation of the burden of Chagas disease and dynamics of Trypanosoma cruzi transmission. Methodology/Principal Findings: We conducted a cross-sectional serological and entomological study to evaluate temporal and spatial patterns of T. cruzi transmission in a peri-rural region of La Joya, Peru. We use a multivariate catalytic model and Bayesian methods to estimate incidence of infection over time and thereby elucidate the complex history of transmission in the area. Of 1,333 study participants, 101 (7.6%; 95% CI: 6.2–9.0%) were confirmed T. cruzi seropositive. Spatial clustering of parasitic infection was found in vector insects, but not in human cases. Expanded catalytic models suggest that transmission was interrupted in the study area in 1996 (95% credible interval: 1991–2000), with a resultant decline in the average annual incidence of infection from 0.9% (95% credible interval: 0.6–1.3%) to 0.1% (95% credible interval: 0.005–0.3%). Through a search of archival newspaper reports, we uncovered documentation of a 1995 vector control campaign, and thereby independently validated the model estimates. Conclusions/Significance: High levels of T. cruzi transmission had been ongoing in peri-rural La Joya prior to interruption of parasite transmission through a little-documented vector control campaign in 1995. Despite the efficacy of the 1995 control campaign, T. cruzi was rapidly reemerging in vector populations in La Joya, emphasizing the need for continuing surveillance and control at the rural-urban interface.
Tamerius, J.D. and Comrie, A.C., 2011: Coccidioidomycosis incidence in Arizona predicted by seasonal precipitation. PLoS ONE 6(6): e21009. doi:10.1371/journal.pone.0021009.
The environmental mechanisms that determine the inter-annual and seasonal variability in incidence of coccidioidomycosis are unclear. In this study, we use Arizona coccidioidomycosis case data for 1995–2006 to generate a timeseries of monthly estimates of exposure rates in Maricopa County, AZ and Pima County, AZ. We reveal a seasonal autocorrelation structure for exposure rates in both Maricopa County and Pima County which indicates that exposure rates are strongly related from the fall to the spring. An abrupt end to this autocorrelation relationship occurs near the the onset of the summer precipitation season and increasing exposure rates related to the subsequent season. The identification of the autocorrelation structure enabled us to construct a “primary” exposure season that spans August-March and a “secondary” season that spans April–June which are then used in subsequent analyses. We show that October–December precipitation is positively associated with rates of exposure for the primary exposure season in both Maricopa County (R = 0.72, p = 0.012) and Pima County (R = 0.69, p = 0.019). In addition, exposure rates during the primary exposure seasons are negatively associated with concurrent precipitation in Maricopa (R = −0.79, p = 0.004) and Pima (R = −0.64, p = 0.019), possibly due to reduced spore dispersion. These associations enabled the generation of models to estimate exposure rates for the primary exposure season. The models explain 69% (p = 0.009) and 54% (p = 0.045) of the variance in the study period for Maricopa and Pima counties, respectively. We did not find any significant predictors for exposure rates during the secondary season. This study builds on previous studies examining the causes of temporal fluctuations in coccidioidomycosis, and corroborates the “grow and blow” hypothesis.
Uejio, C.K., Kemp, A. and Comrie, A.C. 2012: Climatic controls on West Nile virus and Sindbis virus transmission and outbreaks in South Africa. Vector-Borne and Zoonotic Diseases 12(2): 117-125. doi:10.1089/vbz.2011.0655.
The processes influencing the magnitude of West Nile virus transmission from one year to the next require thorough investigation. The intensity of West Nile virus transmission is related to the dynamics and interactions between the pathogen, vector, vertebrate hosts, and environment. Climatic variability is one process that can influence inter-annual disease transmission. South Africa has a long West Nile virus and Sindbis virus record where consistent climate and disease relationships can be identified. We relate climate conditions to historic mosquito infection rates. Next, we detect similar associations with reported human outbreaks dating back to 1941. Both concurrent summer precipitation and the change in summer precipitation from the previous to the current summer were strongly associated with West Nile and Sindbis virus transmission and recorded human outbreaks. Each 100mm change in inter-annual summer precipitation change increased WNV infection rates by 0.39 WNV positive Cx. univittatus/1000 tested Cx. univittatus. An improved understanding of biotic and abiotic disease transmission dynamics may help anticipate and mitigate future outbreaks.
Stacy, P.K.R., Comrie, A.C. and Yool, S.R. 2012: Modeling valley fever incidence in Arizona using a satellite-derived soil moisture proxy. GIScience & Remote Sensing 49, 299–316, doi:10.2747/1548-1603.49.2.299.
Valley Fever is caused by inhalation of spores from the soil-dwelling fungus Coccidioides spp. Pima, Pinal and Maricopa counties, Arizona, have the highest Valley Fever incidence on earth. Despite reported links between climate, habitat, disease timing and seasonality, relationships between the fungus and its putative affinity to moist soils are poorly understood. We used Normalized Difference Vegetation Index time series from the Advanced Very High Resolution Radiometer to compare soil moisture variations with disease incidence. Results suggest moist soils in the early spring, resulting from antecedent winter precipitation, correlate with increased incidence in these counties up to a year later.
Scott, C.A., Robbins, P.F. and Comrie, A.C., 2012: The mutual conditioning of humans and -pathogens: implications for integrative geographical scholarship. Annals of the Association of American Geographers 102 (5), 977-985.
We highlight an emerging mode of human-environment enquiry that is executed by cross-disciplinary teams, spurs innovation of hybrid methods, and leads to non-intuitive findings relevant beyond disciplinary framings or specific cases. The extension of this approach in health geography is particularly instructive. By focusing on material objects like soils, insects, or sewage, researchers from diverse epistemologies are compelled to translate conceptual models of disease causation, risk, and vulnerability. Humans and pathogens mutually condition one another, a result of continuously changing exposures (settlement and development patterns that modify pathogen and vector ecology) and institutional processes (legal, economic, and organizational contexts in which environments are modified and agents respond to risk). The dynamic interactions of pathogen ecologies and human institutions produces a type of coevolution, as evidenced by three cases we consider: 1) bacteriological and helminth infections from urban wastewater irrigation, 2) West Nile Virus and its mosquito vector in the built environment, and 3) Valley Fever and fungal distribution under changing climate and land disturbance. Place-based, contextual exposure pathways are shown to provide only partial explanation of disease transmission and must be complemented by insights into individual and organizational agents’ motivations, logics, and responses. The object in its context holds the key to understanding the intersection between physical/environmental and human/governance geographies. Interactively identifying and pursuing theoretical and applied challenges in this manner allows researchers to move beyond entrenched sub-disciplinary understandings to frame new supra-disciplinary questions.
Comrie, A.C. and McCabe, G.J., 2012: Global air temperature variability independent of sea-surface temperature influences. Progress in Physical Geography 37, 29-35, doi: 10.1177/0309133312460074.
Mean global surface air temperature (SAT) and sea surface temperature (SST) display substantial variability on time scales ranging from annual to multi-decadal. We review the key recent literature on connections between global SAT and SST variability. Although individual ocean influences on SAT have been recognized, the combined contributions of worldwide SST variability on the global SAT signal have not been clearly identified in observed data. We analyze these relations using principal components of detrended SST, and find that removing the underlying combined annual, decadal, and multi-decadal SST variability from the SAT time series reveals a nearly monotonic global warming trend in SAT since about 1900.
El Vilaly, A.E., Arora, M., Butterworth, M.K., El Vilaly, M.A.M., Jarnagin, W. and Comrie, A.C., 2012: Climate, environment and disease: The case of Rift Valley fever. Progress in Physical Geography, online, doi: 10.1177/0309133313478315.
Rift Valley fever is a disease of animals and humans found throughout much of Africa, and recently in the Arabian Peninsula. It is spread via mosquito vectors and direct contact with infected tissue and fluids. Climate variability and change alter ecological processes involved in the outbreak and spread of diseases such as Rift Valley fever. This progress report reviews the key research literature on climate-driven environmental change and Rift Valley fever. The roles of regional and seasonal climates for the disease are emphasized, as well as remote sensing and other approaches to monitoring and analysis. The paper concludes with five suggested future directions for research.
Tamerius J.D., Shaman, J., Alonso, W.J., Bloom-Feshbach, K., Uejio, C., Comrie, A.C. and Viboud, C., 2013: Environmental predictors of seasonal influenza epidemics across temperate and tropical climates. PLoS Pathogens 9(3): e1003194. doi:10.1371/journal.ppat.1003194.
Human influenza infections are strongly seasonal in temperate regions. Recent laboratory and epidemiological evidence suggests that low specific humidity conditions facilitates the airborne survival and transmission of the influenza in temperate regions, resulting in annual winter epidemics. However, this is unlikely to account for the epidemiology of influenza in tropical and subtropical sites where epidemics often occur during the rainy season or transmit year-round without a well defined season. We assessed the role of specific humidity and other local climatic variables on influenza virus seasonality by modeling epidemiological and climatic information from 78 study sites sampled globally. We substantiated that there are two types of environmental conditions associated with seasonal influenza epidemics: “cold-dry” and “humid-rainy”. For sites where monthly average specific humidity or temperature decreases below thresholds of approximately 11-12 g/kg and 18-21 ºC during the year, influenza activity peaks during the cold-dry season (i.e., winter) when specific humidity and temperature are at minimal levels. For sites where specific humidity and temperature do not decrease below these thresholds, seasonal influenza activity tends to peak in months when average precipitation totals are maximal and greater than 150 mm per month. These findings provide a simple climate-based model rooted in empirical data that accounts for the diversity of seasonal influenza patterns observed across temperate, subtropical and tropical climates.