Final Report: Response of Regional Air Quality to Severe Drought

EPA Grant Number: R835191
Title: Response of Regional Air Quality to Severe Drought
Investigators: Allen, David T. , Huang, Ling , Kimura, Yosuke , McDonald-Buller, Elena , McGaughey, Gary , Zheng, Jeff
Institution: The University of Texas at Austin
EPA Project Officer: Chung, Serena
Project Period: June 1, 2012 through May 31, 2015 (Extended to May 31, 2016)
Project Amount: $750,000
RFA: Extreme Event Impacts on Air Quality and Water Quality with a Changing Global Climate (2011) RFA Text |  Recipients Lists
Research Category: Air Quality and Air Toxics , Global Climate Change , Water and Watersheds , Climate Change , Air , Water

Objective:

Drought is a natural disaster that has profound and complex social, economic, and environmental impacts. In recent years, the severity of drought has varied spatially and temporally in Texas, which has highly diverse climatic conditions and land use and land cover profiles over its ten climate regions. The National Climate Assessment (Melillo, et al., 2014) indicates an increase in the state over the next several decades in the annual numbers of consecutive dry days and numbers of days with temperatures exceeding 100°F. Changes in vegetation associated with more frequent droughts have the potential to affect regional air quality, which will be critical to understand as the state endeavors to achieve and maintain attainment with National Ambient Air Quality Standards (NAAQS) for ozone. This study characterized variations in eastern Texas land cover, biogenic emissions, meteorological conditions, and dry deposition, individually as well as simultaneously, during representative drought and average-towet periods. Specific objectives included (1) investigating annual and seasonal variations in leaf area index (LAI), soil moisture, and meteorological variables on emissions estimates of biogenic hydrocarbons using the Model of Emissions of Gases and Aerosols from Nature (MEGAN); (2) characterizing differences between a regional land cover product with high spatial resolution and a widely used global land cover product on biogenic emissions estimates and ozone predictions; (3) investigating the spatial and interannual variability of the most common crops planted and harvested in Texas; (4) examining seasonal and interannual changes in estimated ozone dry deposition velocities and component resistances over eastern Texas using a dry deposition algorithm from the Comprehensive Air Quality Model with Extensions (CAMx); and (5) examining the relative contributions of different physical and chemical processes on predicted ground-level ozone concentrations using the process analysis diagnostic tool in CAMx during representative drought and wet periods. 

Summary/Accomplishments (Outputs/Outcomes):

Eastern Texas contains most of the metropolitan areas of the state, including urban areas that are currently in nonattainment or facing that designation under the 8-hour NAAQS for ozone. Biogenic emissions represent a significant fraction of the volatile organic compound (VOC) inventories in eastern Texas ozone nonattainment areas, and consequently, have a critical role in air quality planning efforts. Our analyses used ground- and satellite-based observations and employed widely-applied global and regional air quality models to quantify environmental differences between years with severe to extreme drought (e.g., 2006 and 2011) and years with average to wet precipitation patterns (e.g., 2007 and 2010). Key findings are summarized below: 
  • A pathway through which drought can affect estimates of emissions of isoprene and other biogenic volatile organic compounds is through changes in leaf area index (LAI), a key input parameter for biogenic emissions models. Maximum monthly interannual LAI variations during 2006-2011 exceeded 20% in climate regions with low-growing (i.e., grasses and crops) vegetation (North and South Central Texas) but were less than 20% in heavily forested (East Texas) areas. Drought-induced reductions in LAI values were estimated to result in lower isoprene and monoterpene emissions in central Texas by as much as -24%; maximum interannual variability in monthly isoprene emissions exceeded 30%. 

  • Variability in environmental inputs that influence isoprene and monoterpene emissions within eastern Texas was quantified by examining seasonal and interannual changes in activity factors intrinsic to the Model of Emissions of Gases and Aerosols from Nature (MEGAN). Temperature was found to be the primary driver of variations in isoprene and monoterpene emissions; during drought years, reductions in LAI were dominated by predicted emissions increases caused by much warmer temperatures. The response of biogenic emissions to reduced soil moisture availability to vegetation is a major source of uncertainty; dependent on the specific soil moisture database employed, predicted reductions in isoprene emissions ranged from minimal to -70% during the summer of 2011. These findings emphasize a continued need for investigations to evaluate and improve the water stress parameterizations (both short- and long-term) and/or representations in models such as MEGAN. 

  • Soil moisture is the primary mechanism by which drought effects are manifested in isoprene emission estimates in biogenic estimation models such as MEGAN. The input soil moisture databases typically employ predictions from land surface models (LSMs); however, the lack of in-situ monitoring locations hampers the capabilities to incorporate the observational data as well as to validate the regional soil moisture simulations. Our results demonstrated a high temporal and spatial sensitivity to both the absolute water contents and wilting point values. Future work that generates inputs and/or evaluates outputs from LSMs with additional in-situ monitoring as well as comparisons to satellite-derived (e.g., NASA’s Soil Moisture Active Passive or SMAP) soil moisture estimates would be beneficial. 

  • Although the LSM datasets used in our study had a dry bias with respect to deep soil moisture, predicted isoprene emissions were relatively insensitive to deep soil moisture because a relatively low proportion of root mass was estimated at depths of 40-200 cm. These results emphasize the need for an accurate representation of the vertical soil structure and distribution of root mass of biogenic emitting species and suggests the importance of additional analysis of regionally-representative root distributions. 

  • Uncertainties in land cover characterizations affect modeled biogenic emissions, primarily associated with the standard emission potentials that are dependent on land cover distributions. Misclassification between trees and grasses/crops between global and regional land cover products resulted in substantially different biogenic emissions estimates (by as much as a factor of 10) and predicted ground-level ozone concentrations (mean differences of 2-6 ppb in maximum daily 8-hour average ozone concentrations) for eastern Texas. Overall, these findings indicated that the uncertainties associated with land cover data could lead to uncertainties in modeled biogenic emissions that could be even greater than using different biogenic emission models. These results also highlight the particular importance to air quality in other regions of the world where rapid land cover change, such as deforestation, is occurring. 

  • An investigation of available in-situ ground-based monitoring data in eastern Texas during 2006-2014 indicated minimum and maximum isoprene concentrations during winter and summer, respectively; spring and fall had generally comparable values. The substantial variability in concentrations among monitoring locations emphasized the importance of localized emissions and/or chemistry. Measurements at a limited number of monitors surrounded by undeveloped land and/or natural vegetation are consistent with the hypothesis that biogenic emissions are reduced during drought compared to nondrought periods. Predicted isoprene concentrations obtained from photochemical modeling were often substantially greater than the available in-situ measured concentrations. Emissions estimates from MEGAN are influenced by environmental input factors that may have different directional changes in response to climatic conditions. During drought years, isoprene emissions estimates were influenced by the competing effects of decreases in LAI, increases in leaf age, PAR, and especially temperature, as well as the selection of soil moisture and wilting point databases. However, the persistent, large spatial scale of overprediction of isoprene concentrations during both drought and nondrought periods suggests that investigation of the gridded basal emission factors used to estimate isoprene emissions within MEGAN is warranted. 

  • The temporal and spatial patterns of crops grown in Texas were investigated using information provided by the National Agricultural Statistics Service (NASS). Average statewide planted and harvested acreages during 2006-2013 were 17.51 million acres and 16.34 million acres, respectively; the most densely planted agricultural districts (> 25% areal coverage) are in portions of southern and northwestern Texas with minimal activity in eastern and southwestern Texas. The most common crops planted are cotton (33.5% of the statewide area planted), wheat (33.5%), sorghum (14.0%), and corn (12.2%). The minimum to maximum differences of statewide planted acreages during 2006-2013 for wheat and corn was < 35% compared to more than a factor of two difference for sorghum. Almost half of crops grown in the High Plains are irrigated whereas non-irrigation dominates most of the remaining districts. Wheat and, more markedly, cotton showed evidence that irrigation substantially improved crop harvests, highlighting the potential importance of irrigation practices on variations in land cover. 

  • Average annual planted acreage estimates based on the Cropland Data Layer (CDL) were compared to those from NASS. For cotton and wheat (and to a lesser extent, corn), the directional interannual trends within the most densely planted agricultural districts were generally consistent between the two datasets; the majority of annual CDL values were +/- 25% of the NASS but maximum differences for any given year and region could exceed 50%. The CDL to NASS comparison for sorghum showed poor agreement. Across all crop types, the CDL values were sometimes greater than NASS and vice versa; a consistent bias was not observed. 

  • A comparison of the harvests between drought and nondrought years indicated that harvested-to-planted percentages were greatest during average-to-wet years for all crop types but had significant variability. Corn had consistently high annual harvested percentages across most years (> 90%) with a minimum value of 82% during the all-time record drought year 2011. Cotton and wheat had substantial year-to-year variability with maximum harvested percentages during average-to-wet years (> 95% for cotton and > 60% for wheat) and substantially lower percentages during drought (< 40% for both crops during 2011). A specific year could have significant regional variability that reflected the variations in drought conditions; for example, severe drought during 2009 in South Central Texas was associated with a harvested percentage for cotton of only 10% compared to > 75% for Low and High Rolling Plains that had normal-to-moderate drought conditions. 

  • Dry deposition represents a critical pathway through which ground-level ozone is removed from the atmosphere. Understanding the effects of drought on ozone dry deposition is essential for air quality modeling and management in regions of the world with recurring droughts. Our study applied the widely used Zhang (2003) dry deposition algorithm to examine seasonal and interannual changes in estimated ozone dry deposition velocities and component resistances/conductances over eastern Texas during years with drought (2006 and 2011) as well as a year with slightly cooler temperatures and above average rainfall (2007). Simulated area-averaged daytime ozone dry deposition velocities ranged between 0.26 and 0.47 cm/s. Seasonal patterns reflected the combined seasonal variations in non-stomatal and stomatal deposition pathways. Daytime ozone dry deposition velocities during the growing season were consistently larger during 2007 compared to 2006 and 2011. These differences were associated with differences in stomatal conductances and were most pronounced in forested areas. Reductions in stomatal conductances under drought conditions were highly sensitive to increases in vapor pressure deficit and warmer temperatures in Zhang's algorithm. Reductions in daytime ozone deposition velocities and deposition mass during drought years were associated with estimates of higher surface ozone concentrations. Results from this study emphasize the strong need for field measurements and the importance of understanding the spatial distribution of impacts on ozone dry deposition over eastern Texas and other regions of the world subject to recurring drought. 

References:

Melillo JM, Richmond TC, Yohe GW, eds. Climate Change Impacts in the United States. Third National Climate Assessment. U.S. Global Change Research Program, 2014. 
 
Zhang L, Brook JR, Vet R. A revised parameterization for gaseous dry deposition in airquality models. Atmospheric Chemistry and Physics 2003;3(6):2067-2082. 


Journal Articles on this Report : 4 Displayed | Download in RIS Format

Other project views: All 11 publications 4 publications in selected types All 4 journal articles
Type Citation Project Document Sources
Journal Article Huang L, McDonald-Buller EC, McGaughey G, Kimura Y, Allen DT. Annual variability in leaf area index and isoprene and monoterpene emissions during drought years in Texas. Atmospheric Environment 2014;92:240-249. R835191 (2013)
R835191 (2014)
R835191 (Final)
  • Full-text: ScienceDirect-Full Text HTML
    Exit
  • Abstract: ScienceDirect-Abstract
    Exit
  • Other: ScienceDirect-Full Text PDF
    Exit
  • Journal Article Huang L, McGaughey G, McDonald-Buller E, Kimura Y, Allen DT. Quantifying regional, seasonal and interannual contributions of environmental factors on isoprene and monoterpene emissions estimates over eastern Texas. Atmospheric Environment 2015;106:120-128. R835191 (2014)
    R835191 (Final)
  • Full-text: ScienceDirect-Full Text PDF
    Exit
  • Abstract: ScienceDirect-Abstract
    Exit
  • Other: ResearchGate-Full Text PDF
    Exit
  • Journal Article Huang L, McDonald-Buller E, McGaughey G, Kimura Y, Allen DT. Comparison of regional and global land cover products and the implications for biogenic emission modeling. Journal of the Air & Waste Management Association 2015;65(10):1194-1205. R835191 (2014)
    R835191 (Final)
  • Abstract from PubMed
  • Full-text: ResearchGate-Prepublication Full Text PDF
    Exit
  • Abstract: Taylor & Francis-Abstract
    Exit
  • Other: Taylor & Francis-Full Text PDF
    Exit
  • Journal Article Huang L, McDonald-Buller EC, McGaughey G, Kimura Y, Allen DT. The impact of drought on ozone dry deposition over eastern Texas. Atmospheric Environment 2016;127:176-186. R835191 (Final)
  • Full-text: ScienceDirect-Full Text HTML
    Exit
  • Abstract: ScienceDirect-Abstract
    Exit
  • Other: ScienceDirect-Full Text PDF
    Exit
  • Supplemental Keywords:

    biogenic emissions, soil moisture, drought, dry deposition, ozone, MEGAN, CAMx 

    Progress and Final Reports:

    Original Abstract
    2012 Progress Report
    2013 Progress Report
    2014 Progress Report