Grantee Research Project Results
2006 Progress Report: Effects of Climate Change on Human Health: Current and Future Impacts
EPA Grant Number: R832751Title: Effects of Climate Change on Human Health: Current and Future Impacts
Investigators: Hanna, Adel , Yeatts, Karin B. , Xiu, Aijun , Henderson, Fred , Robinson, Peter , Smith, Richard , Zhu, Zhengyuan
Institution: University of North Carolina at Chapel Hill
EPA Project Officer: Chung, Serena
Project Period: January 1, 2006 through December 31, 2008 (Extended to December 31, 2010)
Project Period Covered by this Report: January 1, 2006 through December 31, 2007
Project Amount: $599,103
RFA: The Impact of Climate Change & Variability on Human Health (2005) RFA Text | Recipients Lists
Research Category: Climate Change
Objective:
The overall goal of this project is to define more precisely the interrelationships among (a) changes in climate and meteorological conditions, (b) air pollution, and (c) heat- and cold-related morbidity severe enough to warrant clinical contact. A secondary goal is to evaluate heat-related morbidity in a vulnerable population: children and adults under economic disadvantage.
Progress Summary:
We are examining the above interrelationships in 12 cities across North Carolina. The state’s varied terrain—mountain, Piedmont, coastal—makes it a good choice for examining the potential impacts of climate variability and air quality on health. In this first project year we performed our analysis for the Charlotte, NC, area. Ten years of data (1996-2005) were used, including weather observations (daily maximum temperature, daily average wind speed, daily minimum temperature, surface pressure, dew point), air quality measurements (O3, NO2, CO, PM10), and hospital admissions records of asthma and myocardial infarction (MI). Daily weather and climate conditions in the Charlotte area are classified in terms of eight air mass types that characterize their origin (tropical or polar) and conditions (moist/dry/cloudy, and temperature). Most air mass types show a seasonal cycle; the moist tropical and moist moderate air masses peak during summer, while the dry polar and dry moderate maximum occurrences fall during autumn and winter.
We found that the moist tropical and dry moderate air mass types are associated with more ozone episodes (periods with high ozone concentrations) than other air masses are. While the dry tropical air mass type comes in a distant third behind the first two types, it is associated with higher peaks of ozone concentrations, indicating more severe ozone episodes than are associated with any of the other air masses. This explains the positive correlations between the lag 1-day ozone concentrations and asthma admissions under dry tropical air masses, as described below. The correlations between ambient ozone concentrations and the daily maximum temperature are positive and significant.
We used a generalized linear model (GLM) to study the regression relationship between 0-, 1- , and 2-day lagged O3, NO2, PM10, and CO concentrations, air mass types, and asthma and myocardial infarction (MI) hospital admissions in adults, after adjusting for meteorological variables, nonlinear seasonal effects, and long-term trend. We assume the number of hospitalizations on each day has a Poisson distribution with mean lambda, and the logarithm of lambda is linearly related to some of the meteorological and air quality variables. Fitting such a GLM allows us to quantify the potential health effects of a given air pollutant after adjusting for meteorological variables. We also include in the model the interaction effect between air mass type and the air quality data in order to assess whether the health effects are dependent on air mass type. The nonlinear seasonal effects and long-term trend are modeled using a B-spline function.
Current-day NO2 was marginally related to asthma hospitalizations under moist tropical air masses (p-value = 0.067) and transitional air masses (p-value = 0.099). We found no significant linear relationship between current-day ozone and asthma admissions under any air mass type. However, lag 1-day ozone concentration is positively related to asthma admissions under dry tropical air masses (p-value = 0.04). No significant linear relationship between lag 2-day ozone and asthma admissions was found under any air mass type. Current-day PM10 positively related to asthma admissions under dry moderate air masses (p-value = 0.025) and moist moderate air masses (p-value = 0.067). Lag 1-day PM10 positively related to asthma hospital admissions under dry tropical air masses (p-value = 0.034). No significant linear relationship between lag 2-day PM10 and asthma admissions was found under any air mass type. All of these results were obtained after adjusting for differences in the daily maximum relative humidity, average daily pressure, departure from normal temperature, average daily dew-point temperature, and average daily wind speed.
Under dry tropical air masses, lag 2-day PM10 was associated with increased MI admissions (p-value < 0.03), while under moist tropical air masses PM10-associated MI admissions were slightly reduced. Maximum daily NO2 concentrations were strongly associated with increased MI admissions (p-value < 0.04), but for moist polar air masses the association was marginally reduced (p-value = 0.059). No statistically significant associations of ozone and MI hospital admissions were found under any air mass type.
We conclude in general that certain synoptic air mass patterns, in conjunction with ambient air pollution levels, are associated with increased asthma and MI hospital admissions. Further analysis is needed to confirm and elaborate on these findings. We believe that additional health data (e.g., office visits) are needed to confirm the conclusions, and are working on getting these data.
Future Activities:
During the project’s second year, we will apply the Charlotte air mass classification system and evaluate the corresponding meteorological, air quality, and health data associations in 11 other cities geographically distributed across North Carolina. We will also initiate the modeling portion of our investigation. We will acquire Community Climate System Model (CCSM) model simulations of 1996 through 2030 for base case and CO2-controlled experiments. We will first examine how well this global model (CCSM) can replicate SSC patterns that have been reported, based on the observational analysis of current years (1996-2005).
Journal Articles:
No journal articles submitted with this report: View all 11 publications for this projectSupplemental Keywords:
global climate, air quality, human health, epidemiology, modeling, climate models, Southeastern U.S.,, RFA, Health, Scientific Discipline, Air, Health Risk Assessment, climate change, Air Pollution Effects, Risk Assessments, Biochemistry, Environmental Monitoring, Ecological Risk Assessment, Atmosphere, air quality modeling, morbidity, air pollution, human exposure, climate models, human dimension, human health risk, land use, statistical methodsRelevant Websites:
http://cf.unc.edu/cep/empd/projects2/climate/index.cfm Exit (note that the site is password protected and, for now, can be accessed only by the team of investigators and the EPA Project Officer).
Progress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.