Skip to main content

A cross-divisional department spanning

Research and Practice

2023 EHE Research Retreat

Department of Environmental Health &
Engineering | 2023 Annual Research Retreat

Mount Washington Conference Center
5801 Smith Avenue
Baltimore, MD 21209

*Masks will be provided and wearing them when not eating or drinking is strongly encouraged.
* We encourage everyone to bring a reusable bottle to take advantage of the water filtration stations.


2023 Research Retreat AGENDA

DATE: Friday, January 20, 2023
TIME: 8:00am – 4:00pm followed by poster awards and reception (4:00pm – 5:30pm)

DURATION: 8 hours

Check-in, Breakfast, and Poster Setup

Time: 8:00am – 9:00am      
Duration: 60 minutes 

     Check-in: Main entrance of Conference Center
     Breakfast: Chesapeake Galley (can also eat outside in North or South Dining Terrace)
     Poster setup: Breakout rooms 200 & 202 (on second level of facility)

Introduction from Chair 

Time: 9:00am – 9:15am      
Duration: 15 minutes 
Where: Pullen Plaza

Ice Breakers

Time: 9:15am – 9:25am 
Duration: 10 minutes 
Where: Pullen Plaza


Poster Session #1 (evens)

Time: 9:30am – 10:30am
Duration: 60 minutes
Where: Breakout rooms 200 and 202


New Faculty Introductions & Presentations

Time: 10:40am – 12:00pm 
Duration: 80 minutes 
Where: Pullen Plaza


Time: 12:00pm – 1:00pm 
Duration: 60 minutes 
Where: Chesapeake Galley (can also eat outside in North or South Dining Terrace)


Poster Session #2 (odds)*

Time: 1:15pm – 2:15pm
Duration: 60 minutes
Where: Breakout rooms 200 and 202


Presentations from Centers in EHE

Time: 2:25pm – 3:35pm
Duration: 70 minutes
Where: Pullen Plaza

Presentations from Centers in EHE: Panel Q&A

Time: 3:35pm – 4:00pm
Duration: 25 minutes
Where: Pullen Plaza

Wrap-up, Poster Awards & Reception

Time: 4:00pm – 5:30pm
Duration: 90 minutes
Where: Chesapeake Galley (can also eat outside in North or South Dining Terrace)

*All poster presenters should take down their posters at this time; they must be collected no later than 3pm.

Public Health Abstracts

AlamElDin, Sarah Marie

Cellular Metal Transport from Electronic Cigarette Exposure  

Background:  The recent rapid rise in electronic cigarette (e-cig) use, is of concern as several studies have shown that e-cig can induce cell death, oxidative stress, oxidative damage, and altered metabolic activity.  However, which specific components and mixtures in e-cig aerosols cause these health effects is still unknown.  Heavy metals as well as nanoparticles have been found in electronic cigarette aerosols and are a potential source of toxicity. Additionally, concentrations of metals such as Ni and Cr have been reported in urine of users at concentrations that correlate with use of e-cig devices.  

Objective/aim:  This observation supports the hypothesis that metals in e-cig aerosols are inhaled and transported across the lung epithelial barrier in e-cig users. The specific mechanism of how metals can be taken up and transported by lung cells is currently unknown. Physical and chemical properties of the metals can affect this mechanism. For example, it is known that Ni nanoparticles are taken up by cells while NiCl2 is poorly absorbed by cells. Additionally, Cr (VI) is a known inhalation carcinogen while Cr (III) is thought to be an essential metal.  

Methods: Here we investigate metal uptake in A549 lung epithelial cells exposed to NiCl2, Ni nanoparticles, CrCl3 (Cr (III)), K2Cr2O7 (Cr (VI)), and Cr2O3 (Cr (III)) nanoparticles. After exposure cells were washed with 5mM EDTA and digested using 20% optima nitric acid and analyzed using an Agilent 8900 ICP-MS.  

Results:  Exposure to Ni nanoparticles for 24 hours resulted in a dose response increase in Ni concentrations in cells. Increases in cellular Cr concentrations were not seen after nanoparticle treatment. However, exposure to Cr (VI) resulted in a dose dependent increase in cellular Cr concentrations.  Conclusion: These results indicate that there is some specificity and differences in the mechanism of uptake of Cr and Ni nanoparticles. It is known that surface chemistry can affect the uptake of nanoparticles into cells and are possible reasons for why treatment with Cr (III) nanoparticles did not result in increased cellular Cr (III) concentrations. 

Chiger, Andrea

Investigating the Modifying Effects of Allostatic Load on Associations Between Metals and Blood Pressure  

AUTHORS: Chiger, A.A.; Buckley, J.P.; Jones, M.R.; Kuiper, J.R.; Nachman, K.E.

BACKGROUND: Disadvantaged populations are often co-exposed to high levels of metals and psychological stressors. While limited research suggests that stress may increase susceptibility to lead cardiotoxicity, the influence of stress on other metals remains unexplored. AIM: We examined whether allostatic load (AL), a measure of chronic stress, modifies associations between metals and blood pressure (BP).

METHODS: We analyzed cross-sectional data on blood and urinary metal biomarkers and systolic and diastolic BP in 5,022 adults (20+ years) participating in NHANES 2003-2010 using linear regression models. We assessed whether associations were modified by AL using interaction terms and stratified analyses. Because there is no scientific consensus on how to measure AL, we compared four different approaches to dichotomize participants into high and low AL based on seven stress biomarkers. Final models for each metal were adjusted for age, sex, race/ethnicity, BMI, education level, serum cotinine, urinary creatinine (to account for urinary dilution), and other confounding metals (e.g., lead).

RESULTS: In models for systolic BP, negative interactions between blood lead and AL were observed, which were consistent across AL definition methods. Stratified quantile g-computation showed that the overall effects of metals on systolic BP were not significant in the high AL group. Use of medical cutoffs to define AL tended to show more modifying effects on the associations between metals and systolic BP than those based on distributions of biomarkers in the population. In diastolic BP models, no consistent interactions were observed. Stratified quantile g-computation models of diastolic BP showed significant positive overall effects of metals in the high AL group only.

CONCLUSIONS: AL may modify the association between certain metals and BP in adults. As results may differ by method used to define AL, future research should use multiple approaches to operationalize AL until consensus is reached on the most appropriate definition. 

Deans, Lauren

Community-Driven Assessments of Exposure and Health Impacts of Industrial Air Pollution in South Baltimore

Authors: (to be updated at later time): Lauren Deanes, Matthew Aubourg, Kristoffer Spicer, Caila Wagar, Christopher D. Heaney Affiliations (all): Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland 21205


Background: Environmental injustices threaten the health and quality of life of those in already vulnerable communities. In the United States, industrial facilities have historically perpetuated these injustices through disproportionate and harmful air emissions. The Maryland Department of the Environment has identified 70 potential sources of air pollution in Curtis Bay (located in South Baltimore), including the CSX Coal Terminal in Curtis Bay. Further, Curtis Bay’s community statistical area (Curtis Bay/Brooklyn/Hawkins Point) has a large community of color and a lower median household income and life expectancy relative to Baltimore City. Local community organizations, such as the South Baltimore Community Land Trust (SBCLT) and the Community of Curtis Bay Association (CCBA), have expressed longtime concerns about exposure and health impacts from air pollution industrial sources on their communities.

Objective/aim: We aim to quantify the spatiotemporal distribution of air pollutants in Curtis Bay through community engagement, hyper-local air monitoring, and source apportionment analyses.

Methods: We collaborated with SBCLT, CCBA, and Curtis Bay businesses to establish a network of air quality monitors throughout the Curtis Bay area in May 2022. Collectively, the air monitors in this growing network measure particulate matter (e.g., total suspended particles (TSP), PM10, PM2.5, etc.), black carbon, gases (e.g., carbon monoxide (CO), nitric oxide (NO), nitrogen dioxide (NO2), etc.), and meteorological data (wind direction, wind speed, relative humidity, and temperature). Further, we supplemented these air quality data with visual observation logs of CSX Coal Terminal activity patterns (trains, ships, “coal dust suppression” sprinklers, bulldozers, and visible smoke/dust), coded from trail camera time-lapse videos deployed in Curtis Bay. We performed t-tests of the difference of log10transformed mean air pollutant concentrations during time periods of downwind vs. non-downwind conditions (relative to the terminal) and when coal terminal activities were vs. were not observed.

Results: Preliminary results indicate that concentrations of all measured particulate matter tend to increase when the wind was blowing downwind vs. non-downwind from the coal terminal towards nearby monitors or when there was vs was not visible dust observed on camera. Additionally, when the terminal’s “coal dust suppression” system is engaged, concentrations of larger particles, expected to be those from coal, tend to increase.

Conclusion: The combination of camera and air monitoring data provides a useful tool for better understanding cumulative impacts of air quality in Curtis Bay, a South Baltimore neighborhood exposed to pollutants from various industrial sources.  

DeSantiago, Melissa

A Pilot Study on Exposures to Heavy Metals in an Occupational Chilean Cohort

Melissa DeSantiago, MPH;1 Lesliam Quirós-Alcalá, PhD, MSc;1 Ana M. Rule PhD, MHS;1 Catterina Ferreccio, MD, MPH; 2 Sandra Cortés A., DVM, PhD, MS.2 AFFILIATIONS  1. Johns Hopkins University, Bloomberg School of Public Health, Department of Environmental Health & Engineering, Baltimore, MD, USA  2. School of Public Health, School of Medicine, Pontificia Universidad Católica de Chile, Santiago, Región Metropolitana, Chile 

Background: Chile’s competitive mining and agricultural industries generate toxic waste from mining ores and agrochemicals that include heavy metals previously linked to an array of adverse health effects such as nephrotoxicity. Still, the extent and determinants of exposure to many heavy metals among worker populations in Chile remains unknown.  

Objective: A pilot study was conducted to assess exposure to heavy metals and examine exposure determinants in an occupational population in the Chilean agricultural region of Maule. Spot urine samples were collected from a subset of 467 workers participating in the Maule Cohort (MAUCO) study.  

Methods. Arsenic (As), chromium (Cr), copper (Cu), manganese (Mn), and Lead (Pb) were quantified in urine samples using ICP-MS. Interviewer-administered questionnaires were administered to capture demographic characteristics, including sociodemographic data, general health history, occupation, and overall lifestyle behaviors.  

Results. About half of the participants were males (53%) and worked in agriculture (53%). Preliminary findings from a subset of 430 participants indicate that detection frequencies (DF) for target metals were: As (DF:86%), Cr (DF:19%), Cu (DF:34%), Mn (DF:25%), Pb (DF:2%). Median As concentrations in agricultural workers were almost twice as high as those reported in adults from the U.S. general population (10.10 vs. 6.02 µg/L).

Conclusion. Ongoing analyses to be presented aim to characterize exposure to heavy metals among all 467 participants and to assess their respective determinants in this subsample of workers to identify modifiable exposure risk factors in a vulnerable working population. 

Garbus, Haley

Formaldehyde Dehydrogenase-Derived Formate Contributes to Cardioprotection in the Female Heart

Background: Ischemic heart disease is a leading cause of death in the US, often manifesting as myocardial ischemia/reperfusion (I/R) injury. The risk for disease, however, is lower in premenopausal women vs. age-matched men and postmenopausal women, suggesting an estrogen-dependent protective mechanism. Previously, we found female hearts to contain two-fold more formaldehyde (FA) vs. males, but also higher activity of FA dehydrogenase (FDH), which detoxifies FA to formate. FDH knockout (KO) exacerbates I/R injury in females, potentially from a lack of FDH-derived formate, yet mechanistic data is lacking.  

Objective/aim: The objective of this study is to investigate a role for estrogen in the regulation of FA production and FDH-mediated metabolism of FA to formate and the role of formate in cardioprotection.  

Methods: To determine differences in enzymes that produce FA, western blots were performed in lysates from male, female, and OVX mouse hearts. Male and female hearts were Langendorff-perfused in the presence or absence of 30 uM formate to determine whether formate is protective during I/R injury. Finally, FDH KO hearts were Langendorff-perfused in the presence or absence of 30uM formate.

Results: Serine hydroxymethyltranserfases 1 and 2 were more highly expressed in females vs. males. Lysine-specific demethylase 1 expression was significantly higher in female hearts vs. OVX and males. Formate treatment reduced infarct size in both sexes of wild-type hearts, and this effect was more pronounced and significant in males. More importantly, we found that formate rescued FDH KO female hearts, reducing infarct size following I/R injury.  

Conclusion: Taken together, our findings demonstrate that formate protects against I/R injury and further suggest that FDH-derived formate is a critical component of female-specific cardioprotection. As FDH is essential for female-specific cardioprotection, we will next determine whether formate rescues OVX hearts to determine its therapeutic potential for postmenopausal women.  

Gigot, Caroline

SARS-CoV-2 and other respiratory viruses among industrial livestock operation workers, North Carolina, 2021-2022  

Carolyn Gigot1, Jordan Kuiper1, Nora Pisanic1, Kristoffer Spicer1, Kate Kruczynski1, Devon Hall, Sr.2, Kirsten Koehler1, Christopher D. Heaney1, 3, 4 1. Johns Hopkins University, Department of Environmental Health and Engineering, Baltimore, MD 2. The Rural Empowerment Association for Community Help (REACH), Warsaw, NC 3. Johns Hopkins Bloomberg School of Public Health, Department of Epidemiology, Baltimore, MD 4. Johns Hopkins Bloomberg School of Public Health, Department of International Health, Baltimore, MD  

Background/Aims: Meat and poultry processing plants emerged as centers of COVID-19 outbreaks across the United States early in spring 2020. We aimed to investigate associations between work and respiratory viruses, including COVID-19, in a North Carolina study population predominantly working at or living near industrial livestock operations (ILOs), and to investigate workplace health and safety protections relevant to infectious disease reported by in-person workers.

Methods: We enrolled 279 participants in North Carolina from February 2021 to July 2022, including 77 ILO workers and 75 people working in person in other industries. Participants completed a baseline questionnaire on demographic and job characteristics, COVID-19, and recent symptoms. We also analyzed participant-collected saliva swab samples for IgG antibodies specific to SARS-CoV-2; human coronaviruses 229E, OC43, HKU1, and NL63; and RSV using a multiplex immunoassay based on Luminex technology. We used GEE logistic regression clustered by household to identify associations of ILO work with antibody responses.  

Results: The prevalence of infection-induced SARS-CoV-2 antibody was high in our study population (53%). More ILO workers (66%) had SARS-CoV-2 infection-induced IgG compared to participants working in person in other industries (45%) (prevalence ratio [PR]=2.2, 95% confidence interval [CI]=1.1, 4.4) and participants either not working or working from home (45%) (PR=2.3, 95% CI=1.2, 4.5). Most ILO workers (89%) and other in-person workers (81%) reported face masks were required at the workplace, but other controls were less frequently reported. More ILO workers reported COVID-19 testing available at work (44%) compared to other in-person workers (26%), but fewer ILO workers reported any changes in absence policies (18%) compared to other in-person workers (26%).  

Conclusion: Questionnaire results suggest low rates of the most effective COVID-19 and infectious disease workplace controls, which may explain some of the relatively high rates of SARS-CoV-2 infection among ILO workers.  

Grace, Daisy

Development of a Novel Analytical Platform for Detection & Identification of Organic Electrophiles in Environmental Matrices 

Background:  Water disinfection processes, including chlorination, have prevented countless cases of waterborne illnesses. An unintended consequence of these disinfectants, however, have resulted in the formation of toxic disinfection byproducts (DBPs). A certain class of DBPs known as “organic electrophiles” exhibit a wide range of toxicity due to their interactions with biological nucleophiles, e.g., DNA base pairs and proteins. Despite their toxicity, organic electrophile DBPs are poorly characterized due to their incompatibility with traditional extraction and detection methods.  

Objective/aim: A novel reactivity-directed analysis approach was developed to simultaneously capture, detect, and identify organic electrophiles formed during water disinfection processes.

Methods:  Nucleophile probes, including biologically relevant thiols and amines, were synthesized onto photocleavable microbeads. The effectiveness of these nucleophile microbeads to enable electrophile capture and detection was evaluated by spiking in standards of organic electrophiles with varying reactivities. The resulting nucleophile-electrophile adducts were subsequently cleaved from the microbeads to analyze via liquid-chromatography-high-resolution mass spectrometry.  

Results:  Thiol microbeads enabled detection of all twelve halogenated organic electrophiles tested in the spiked standard mixture, thus highlighting the suitability of this derivatization method for detecting organic electrophiles formed during chlorination disinfection. No halogenated electrophile adducts were detected using the amine microbeads, thus classifying them as a poor derivatization reagent for electrophiles of this reactivity class. Separately, an organic electrophile mixture containing carbonyls—compounds commonly formed during ozonation— was tested against the amine microbeads. The addition of a reducing agent in solution, or the introduction of an oxygen atom adjacent to the amine nucleophile probe (i.e., an aminooxy group) offered superior stability of the carbonyl-nucleophile bond and offered more sensitive detection of electrophiles of this reactivity class.

Conclusion:  The effective detection of organic electrophiles achieved by the developed microbead assay shows promise in upcoming applications regarding the detection of electrophiles in more complex environmental matrices.  

Kurowski, Kathleen

Source tracking swine fecal waste on household surfaces proximal to swine concentrated animal feeding operations in southeastern North Carolina, USA

Kathleen Kurowski, Kristoffer Spicer, Carolyn Gigot, Nora Pisanic, Devon Hall, Devon Hall Jr, Margaret Carr, Arika Miller, Unique Hall, Kate Kruczynski, Ana M. Rule, Christopher D. Heaney

BACKGROUND AND AIM: In North Carolina, swine industrial livestock operations (ILOs) are located disproportionately in low-income and communities of color. Swine ILO neighbors have raised concerns about emissions from ILOs. We aim to compare the detection frequency and quantity of a swine-specific Bacteroides fecal microbial source-tracking marker (pig-2-bac DNA) contamination on outdoor and indoor household surfaces at ILO worker (ILO) compared to ILO neighbor (ILO-N) and metropolitan resident (metro) homes in North Carolina.

METHODS: At enrollment, ILO, ILO-N, and metro participants collected 2 outdoor and 2 indoor household surface settled dust samples wiping a Copan flocked swab across a 10 cm2 area. Using quantitative real-time PCR, pig-2-bac DNA copy number per area was estimated in household samples. We used logistic and linear regression to evaluate differences in pig-2-bac prevalence and quantity between the three household groups and on outdoor compared to indoor surfaces.

RESULTS: Ninety-one households (354 swabs) were sampled: 30 ILO (113 swabs), 31 ILO-N (119 swabs) and 30 metro (122 swabs). Prevalence of Pig-2-bac detection on >=1 household surface was 33%, 39%, and 0% among ILO, ILO-N, and metro households, respectively. Compared to metro households, higher log-10 pig-2-bac DNA copy number per 10 cm2 was observed at ILO (beta=0.5; 95% confidence interval [CI]=0.2, 0.7) and ILO-N households (beta=0.2; 95% CI=0.06, 0.4). Compared to indoor surfaces, higher log-10 copy number per 10 cm2S was observed on outdoor surfaces (beta = 0.181, 95% CI=0.02, 0.3).

CONCLUSIONS: Swine-specific fecal contamination was detected at a higher prevalence in ILO-N compared to ILO workers and metro resident homes. Both ILO and ILO-N groups had statistically significant increases in copy number compared to the metro group.

KEYWORDS: microbial contamination, pig-2-bac, industrial livestock operations, concentrated animal feeding operations 

Le, Sophia

Modeling Ambient PM2.5 Concentrations at a High Spatiotemporal Resolution between Urban and Rural Settlements in Uganda

Sophia Le1, Laura Nicolaou1,2,3, William Checkley2,3, Darryn Waugh4, Kirsten Koehler1

1 Department of Environmental Health and Engineering, Bloomberg School of Public Health, Johns Hopkins University, Baltimore, MD, 21205, USA

2 Division of Pulmonary and Critical Care, School of Medicine, Johns Hopkins University, Baltimore, Maryland, USA.

3 Center for Global Non-Communicable Disease Research and Training, Johns Hopkins University, Baltimore, Maryland, USA

4 Department of Earth and Planetary Sciences, Krieger School of Arts and Sciences, Johns Hopkins University, Baltimore, Maryland 21218, USA

The 2019 Global Burden of Disease (GBD) study estimated that exposure to ambient fine particulate matter (PM2.5, airborne particles less than 2.5 micrometer in aerodynamic diameter) was responsible for 4.14 million deaths in 2019, overwhelmingly in low- and middle-income countries (LMICs). Currently, disease assessment in LMICs, including those conducted by the World Health Organization and the GBD, rely on extrapolation of North American, European, and Chinese epidemiologic study results. Lack of reliable PM2.5 exposure estimates is a major hurdle to conducting PM2.5 health effects research, and most highly polluted cities in LMICs cannot afford a network of regulatory ground monitors.

Research in the past decade has shown that satellite remote sensing data, especially aerosol optical depth (AOD), can be a promising solution to this challenge. Various statistical and machine learning algorithms have been developed using satellite observations to estimate PM2.5 concentrations. However, this approach remains limited due to the occurrence of missing and inaccurate data in the presence of cloud contamination. To address these limitations, integrating low-cost air quality sensors (LCAQS) with satellite data can improve the spatial and temporal accuracy of model derived PM2.5 estimates. Most of the existing work with LCAQS and satellite observations has been done in the U.S. at relatively low PM2.5 levels compared to LMICs with higher air pollution levels and higher population density.

In this applied study, we propose to develop a data-driven machine learning model to estimate ambient hourly PM2.5 exposure in Uganda. In addition to advancing our knowledge of the complex spatial and temporal patterns of PM2.5 in Sub-Saharan Africa, our work also aims to demonstrate an important application of fusing air pollution exposure estimates from satellite observations, model simulations, and LCAQS as a surrogate for a traditional air quality network in limited resource settings.

Lee-Masi, Monica

Two-Year Evaluation of Legionella in an Aging Residential Building: Assessment of Multiple Water Treatment Approaches

Monica Lee-Masi1, Natalie G. Exum1, Kellogg J. Schwab1 1Environmental Health and Engineering, Bloomberg School of Public Health, Johns Hopkins University, Baltimore, Maryland, USA  

Aging drinking water infrastructure in the US can cause microbial contamination from biofilm-based pathogens that are often present in premise plumbing. Drinking water quality is further degraded when water stagnates in pipes, allowing for microbial growth and colonization of premise plumbing pathogens, particularly Legionella pneumophila (Lp). Our research assessed multiple water treatment approaches over two years for the potential reduction of Legionella in a three-story, three-riser residential building located within an aging, urban drinking water system. Water samples (n=648) were collected and analyzed from hot and cold-water lines at proximal and distal sites within the three risers following removal of showerheads and aerators. At each sampling location 1L was collected from a first draw (Time=0) and after a five-minute flush (Time =5). Lp was quantified via most probable number (MPN) culture and characterized by subsequent molecular detection. Additional physical and chemical parameters including free chlorine, temperature, pH, total organic carbon, and turbidity were measured at each collection site. Building-level water treatment included single thermal heat shocks in hot water systems, single chemical disinfectant shocks in cold and hot water systems, and continuous low-level chemical disinfection treatment in both hot and cold-water systems. We found that, comparing pre- and post-water samples in a paired t-test, single thermal treatments did not ablate Legionella long term (Heat Shock 1: p=0.05 z= -1.93 Heat Shock 2 : p=0.92 z=0.10) and single chemical treatments resulted in increased Legionella long term (Hyperchlorination: p<0.001 z= -4.29).  Continuous low-level chlorine dioxide treatment in hot and cold-water systems reduced Legionella long-term (p<0.001 z= 4.25). This sustained remediation was only achieved through the development and maintenance of a comprehensive water management plan, which is critical to prevent future Legionella outbreaks and reduce the infection risk of Legionnaires’ Disease.     

Lin, Joyce

Racial differences in toenail metal concentration among Gulf Coast residents

Background: The US Gulf region is heavily reliant on the oil and gas, chemical, and agricultural industries and has some of the highest density of EPA toxic release inventory sites per square mile in the country.

Objectives/Aims: We sought to characterize chronic metal exposure in a cohort of Gulf residents and identify personal and environmental determinants of metal exposure in this population.

Methods: We measured toenail metal concentrations using inductively coupled plasma mass spectrometry (ICP-MS) in 413 non-smoking men from the Gulf Long-term Follow-Up Study enrolled in 2011-2013 following the Deepwater Horizon oil spill. We assessed personal, social, and environmental predictors of toenail metal levels using multivariable linear regression. Personal characteristics were ascertained from study questionnaires. Neighborhood social vulnerability by census block group was assessed using data from the US Census American Community Survey. Toxic Release Inventory (TRI) and National Emissions Inventory (NEI) databases were used to assign ambient metal exposures based on geocoded residential locations.

Results: Participant level of education, house type, and the number of Pb emission sources within 10 km, 5 km, and 3 km of the residence were positively associated with toenail Pb concentration. Black participants, who reported having lower levels of educational attainment and greater residential proximity to TRI sites had significantly lower toenail metal concentrations of toxic (Pb, As, Hg, Al, and Cr) and essential metals (Cu, Fe, Mn, Se, Ni, and Zn) compared to their White counterparts adjusting for age, passive smoke exposure, and state of residence.

Conclusion: Toxic toenail metal concentrations were lower among Black participants, which is unexpected given the greater frequency of Black participants living near emissions sources in areas of higher social vulnerability. These differences were not explained by other social and factors. The role of other dietary or behavioral factors should be considered in future studies.

Keywords: Metals, Environmental Justice

This research was supported by a grant from the U.S. Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health to the Johns Hopkins Education and Research Center for Occupational Safety and Health (award number T42 OH0008428), the NIH Common Fund, and the Intramural Research Program of the NIH, National Institute of Environmental Health Sciences (ZO1 ES 102945)

Malloy, Maureen

Low dose ascorbate sensitizes Glioblastoma stem cells to DNA damaging agents by inducing DNA de-methylation and altering the chromatin landscape  

Background: Glioblastoma (GBM) represents the most aggressive and lethal subtype of gliomas. Genetic, epigenetic, and phenotypic heterogeneity within GBM tumors contribute to therapeutic resistance and tumor recurrence. However, the epigenetic reprogramming events remain elusive. The expression and enzymatic product of Ten-Eleven Translocation (TET) enzymes, which can reverse or modify DNA methylation (5mC) to hydroxymethylation (5hmC), has been shown to decrease in gliomas and is correlated with increasing glioma grade and poor prognosis.

Objective/aim: The objective of this study was to determine if re-establishing TET activity and 5hmC levels in tumor cells using ascorbate (vitamin C) has antitumor effects.

Methods: RT-qPCR and Dot blots were used to analyze changes in gene expression and global DNA methylation. Western blots, MTT assay, and immunohistochemistry were used to analyze changes in proteins, cell viability, and 5hmC and gH2A.x co-localization, respectively. Bisulfite sequencing was used to assess specific loci DNA methylation. Endpoints were assessed in glioma stem cells (GSCs).  

Results: We show that low-dose ascorbate enhances TET activity and elevates global 5hmC levels. These changes in DNA methylation correlate with increased tumor cell death induced by temozolomide and ionizing radiation. We also show that 5hmC increases after treatment with DNA damaging agents and co-localizes with gH2A.x. Lastly, we present a novel mechanism whereby ascorbate increases the histone mark H3K36me3 by upregulating GSC expression of the histone methyltransferase NSD1 via demethylation of its gene promoter region.  

Conclusion: We propose that the enhanced susceptibility of ascorbate-treated tumor cells to DNA damage therapies results from increased H3K36me3-induced euchromatin states that increases genomic susceptibility to DNA damage and from the function of 5hmC as a marker of DNA damage foci. These findings underscore the need to understand epigenetic effects in cancer and indicate ascorbate could serve as an adjunct to standard-of-care GBM treatment by re-sensitizing GSCs to DNA damaging therapies.  

Matthews, Shifali

Equity Considerations for Climate and Health Research: A Review of the Literature on Disparate Health Impacts due to Hurricanes and Structural Determinants of Vulnerability

Background: Understanding vulnerability to climate related disasters, such as hurricanes, is crucial for targeting and identifying mitigation and adaptation measures. However, vulnerability assessments often focus on proximal factors and may obscure underlying drivers of health inequities.

Objective/aim: We sought to describe the existing literature on defining vulnerability to hurricanes and develop a framework to incorporate structural determinants of vulnerability.  

Methods: We conducted a literature search in the National Library of Medicine’s MEDLINE/PubMed database. We identified peer-reviewed, population-based research articles assessing equity or vulnerability within the context of hurricane-related health impacts in the U.S., from January 2000 through October 2022. For each study, we determined whether the approach pre-specified vulnerable populations or characterized vulnerability empirically. We also abstracted data on the types of factors used to assess vulnerability to hurricanes (e.g., demographic, structural, etc.). Finally, we developed a framework to guide research on structural determinants of vulnerability to hurricanes.

Results: The initial search resulted in 631 articles. After completion of the title screening, abstracts and full texts of the remaining articles (n=213) were screened and an additional set (n=9 excluded. For the included studies (n=121), 2) were the majority (n=69) pre-specified vulnerable populations of interest, while the remaining 43% (n=52) empirically derived vulnerability. Among all studies, there were common categories of vulnerability, with some overlap: demographic (e.g., race/ethnicity, age, etc.; n=81), spatial (e.g., census tracts, neighborhoods, etc.; n=26), and health status (n=17). Less common categories were outcomes (e.g., displaced; n=6), structural (n=5), and occupational (n=1). Studies on structural vulnerabilities revealed systemic issues of racism, governance, institutions, and infrastructure deficiencies, forming the basis of our conceptual framework.

Conclusion: Most studies on hurricane vulnerability do not consider systemic, upstream factors that contribute to health inequities. Research to drive systemic interventions requires a shift in focus to structural determinants of vulnerability to climate-related disasters .

Miller, Sharmaine

Small Animal Medicine and Animal Shelter Worker Concerns and Sources Received Regarding Animal Welfare during the COVID-19 Pandemic

Authors: Sharmaine L. Miller, Joyce X. Kwan, Kathryn R. Dalton, Kaitlin B. Waite, and Meghan F. Davis Department of Environmental Health and Engineering, Johns Hopkins University Bloomberg School of Public Health, Baltimore Maryland, USA

Background: The COVID-19 pandemic impacted the operations of animal care services worldwide, including veterinary hospitals and animal shelter, rescue, and control (SRC) organizations. Despite ample literature on the concerns of pet-owners during the pandemic, literature is sparse regarding animal welfare concerns from the perspectives of animal care workers, working on the frontlines during this time. Both SRC and small animal medicine workers care for similar pet populations, especially cats and dogs, and within a community, often the same individual pets. Both points of contact can help educate and support pet-owners in maintaining proper pet health and well-being, and potentially alleviate the need for owners to relinquish their pets.

Objective/aim: The goal of this work is to compare small animal medicine and SRC worker perspectives on pandemic-related animal health and welfare outcomes and sources of information and support during the pandemic

Methods: This study, a secondary analysis of collected survey data from the Hopkins COVET study, compared presence of animal health and welfare concerns, the nature of these concerns, frequency of relevant health and welfare updates received, and specific update sources, between SRC and small animal veterinary medicine workers.

Results: It was found that animal SRC animal care workers had a higher odds of reporting having pandemic animal health and welfare concerns, receiving updates on these topics more often, and reporting having expert knowledge on pandemic health and welfare concerns, compared to small animal veterinary hospital workers. Animal health and welfare concerns reported most often by both animal care worker groups included pet-owner financial difficulties, delayed animal care, increases in pet relinquishment and detrimental pet behavioral changes (e.g., due to owners returning to work).

Conclusion: Overall results demonstrate the potential success of frequent updates fostering expert knowledge in animal care workers, but also highlight the need to improve the distribution of animal health and welfare resources to small animal medicine organizations in the future. By doing so , this may bolster support for pet-owners as they interact with either or both animal care service throughout the pets life, and during future public health crises. 

Moore, Elsie

Helping Local Governments Build Food System Resilience: A Mixed-Methods Embedded Intervention Design 

Elsie Moore,1 Meg Burke,2 Karen Bassarab,3 Caitlin Misiaszak,3 Roni Neff,1,3  1. Johns Hopkins Bloomberg School of Public Health  2. Bloomberg Center for Government Excellence  3. Johns Hopkins Center for a Livable Future

Background: Food systems are at risk to a variety of shocks and stressors stemming from human and natural causes. As highlighted by the COVID-19 pandemic, extreme weather events, and the Russian invasion of Ukraine, local and global disruptions impact food security, food supply chains, and food prices. Local governments in the United States play an important role in responding to these food system disruptions.  

Objective: The primary goal of this project was to build the capacity of local governments to prepare for, respond to, and recover from food system disruptions. As this project coincided with the COVID-19 pandemic, we added the aim of evaluating what factors influenced the ease and pace by which local governments responded to the food crises associated with COVID-19. 

Methods:  We recruited participants from five cities in the U.S. to join the Food System Resilience Community of Practice. Through a year-long series of facilitated conversations, city representatives and Johns Hopkins researchers co-developed food system resilience resources for local governments. We used a mixed-method embedded intervention design to collect the Community of Practice and COVID-19 data. We collected pre-intervention survey data in April 2020 and post-intervention survey and interview data in December 2020 and January 2021.  

Results: In December 2022, we released Food System Resilience: A Planning Guide for Local Governments. The six modules and accompanying tools of the Planning Guide take users through a process that results in a set of food system resilience strategies and considerations for implementing them. We found that six key resilience attributes (preparedness, connectivity, redundancy, flexibility, diversity, and capital reserves) played an important role in COVID-19 emergency response efforts.  

Conclusion:  While food system resilience work is nascent, our Planning Guide and COVID-19 research help local governments be better equipped for future disruptive events. 

Moynihan, Emma

Evaluating Ambient Temperature as a Risk Factor for Kidney Disease in the Agricultural Health Study

Background: The National Institute of Environmental Health Sciences and National Cancer Institute’s Agricultural Health Study (AHS) is a prospective cohort study consisting of 52,394 pesticide applicators in Iowa and North Carolina. Enrollment began in 1993 and follow-up is ongoing. AHS's main objective is to understand how environmental exposures and lifestyle factors influence the long-term health of America's agricultural communities.

Objective/Aims: The objective of this poster is to assess whether an association exists between EndStage Kidney Disease (ESKD) and high ambient temperature as a proxy for heat stress. We factor each applicator’s personal protective equipment (PPE) use into their estimated exposure to daily average wet bulb globe temperatures (WBGT) exceeding various occupational recommendations (i.e.., 25 °C).

Methods: We employ a mixed methods approach combining remote sensing data and survey research to estimate each applicator’s occupational exposure to WBGT exceeding recommended thresholds. Cox Proportional Hazard models adjusting for state are implemented. We also evaluate for effect measure modification and interaction with comorbidities, substance use, and occupational tasks.  

Results: Our results indicate chronic exposure too temperature in excess of 25 °C increases the risk of ESKD [Hazard Ratio 1.05, 95% Confidence Interval: (1.00-1.10)] among pesticide applicators in the AHS and among North Carolina applicators in particular HR 1.06 95% C.I: (1.00-1.11)]. We did not observe an effect when restricting analysis to Iowa applicators.

Conclusions: Exposure to temperatures in excess of 25 °C may be associated with an increased risk of ESKD among pesticide applicators in the Agricultural Heath Study. This finding is most pronounced among North Carolina applicators.  

Sola, Steven

Deconstructing WASH: Holistic Approaches to Current and Future Challenges

Background: The United Nations established the 2030 Agenda for Sustainable Development in 2015, which included 17 Sustainable Development Goals (SDGs). SDG6 is specifically focused on the availability and sustainable management of water and sanitation. Climate change is responsible for shifting weather patterns throughout the world. Temperature and rainfall patterns are expected to change how people meet their needs for clean water, increase transmission of waterborne pathogens, and shift how people defecate. how climate change is expected to change There needs to be a greater focus on water use worldwide if the United Nations wants to meet its goal.

Objective/aim: This research has three primary aims change metrics on water collection practices in sub. Aim 1 is to examine the impact of climate Saharan Africa, with an emphasis on vulnerable populations, such as women and girls. Aim 2 is to assess the impact on diarrheal prevalence by household ownership of livestock and other animals in subSaharan Africa. Aim 3  is to determine the risk factors of open-defecation behaviors in rural Odisha, India.

Methods:  This project will use data from the Demographic and Health Survey program, from USAID. These surveys are considered representative from each nation at each time point. Additional data will come from an intervention trial from Caruso et al. (2022) that researched a low-cost, behavior-intervention change intervention to increase latrine use and child feces disposal.

Conclusion:  The goal of this research is to better understand current challenges in the WASH sector, so that future research and interventions are better designed to improve human health. Climate change will continue to adversely affect the water supply around the world; it's important to start addressing challenges now so we’re better prepared  for the future.

Lee-Masi, Monica

Two-Year Evaluation of Legionella in an Aging Residential Building: Assessment of Multiple Water Treatment Approaches 

Monica Lee-Masi1, Natalie G. Exum1, Kellogg J. Schwab1 1Environmental Health and Engineering, Bloomberg School of Public Health, Johns Hopkins University, Baltimore, Maryland, USA  

Aging drinking water infrastructure in the US can cause microbial contamination from biofilm-based pathogens that are often present in premise plumbing. Drinking water quality is further degraded when water stagnates in pipes, allowing for microbial growth and colonization of premise plumbing pathogens, particularly Legionella pneumophila (Lp). Our research assessed multiple water treatment approaches over two years for the potential reduction of Legionella in a three-story, three-riser residential building located within an aging, urban drinking water system. Water samples (n=648) were collected and analyzed from hot and cold-water lines at proximal and distal sites within the three risers following removal of showerheads and aerators. At each sampling location 1L was collected from a first draw (Time=0) and after a five-minute flush (Time =5). Lp was quantified via most probable number (MPN) culture and characterized by subsequent molecular detection. Additional physical and chemical parameters including free chlorine, temperature, pH, total organic carbon, and turbidity were measured at each collection site.  Building-level water treatment included single thermal heat shocks in hot water systems, single chemical disinfectant shocks in cold and hot water systems, and continuous low-level chemical disinfection treatment in both hot and cold-water systems. We found that, comparing pre- and post-water samples in a paired t-test, single thermal treatments did not ablate Legionella long term (Heat Shock 1: p=0.05 z= -1.93 Heat Shock 2 : p=0.92 z=0.10) and single chemical treatments resulted in increased Legionella long term (Hyperchlorination: p<0.001 z= -4.29).  Continuous low-level chlorine dioxide treatment in hot and cold-water systems reduced Legionella long-term (p<0.001 z= 4.25). This sustained remediation was only achieved through the development and maintenance of a comprehensive water management plan, which is critical to prevent future Legionella outbreaks and reduce the infection risk of Legionnaires’ Disease.     

Taube, Nicole

Exposure to Inorganic Arsenic During Pregnancy Alters Structure and Function of the Maternal Heart 

N. Taube, O. V. Ebenebe, B. Lin, R. Kabir, H. Garbus, N. Wang, M. J. Kohr  

Inorganic arsenic (iAs) is a drinking water contaminant in many countries throughout the world, including the U.S. The World Health Organization recommends a limit of 10µg/L iAs in drinking water, yet estimates suggest that 140 million people worldwide consume water with iAs above this threshold. The purpose of this study was to examine the effect(s) of iAs exposure on the maternal heart during pregnancy, with a focus on physiological and molecular changes. We sought to examine the impact of iAs on the maternal heart as this is a relatively understudied area and can highlight potential hazards for pregnant mothers. Pregnant female C57BL/6J mice were exposed to either 1000, 100, or 0µg/L iAs after conception at embryonic day 2.5 with exposure ending at parturition. Dams were harvested at postnatal day 12, and while we previously found that the iAs-exposed pregnant dams did not show typical pregnancy-induced hypertrophy compared to controls, the postpartum maternal heart was enlarged compared to controls as measured by heart weight/tibia length. Transthoracic echocardiography revealed that postpartum mice exposed to 100 or 1000 µg/L iAs had increased ejection fraction, fractional shortening, and cardiac output compared to controls. We next examined contractility in isolated cardiomyocytes from postpartum dam hearts using the Ionoptix system. We found that isolated cardiomyocytes from iAs-exposed dams showed a significant decrease in calcium transient amplitude and sarcomere shortening, along with impaired relaxation kinetics, which contrasts with the enhanced contractile parameters noted in vivo with echocardiography. Together, our findings suggest that exposure to iAs during pregnancy affects postpartum remodeling in the maternal heart, with effects noted at the level of the cardiomyocyte. Furthermore, these findings underscore the detrimental impact of iAs on the cardiovascular system, and specifically highlight potential hazards posed by iAs exposure during pregnancy.

Supported by NHLBI T32ES007141 (NT) and RO1HL13649 (MK). 

Tillery, Anna

Impact of e-cigarette use behaviors and device characteristics on heavy metal biomarkers in a Maryland cohort  

Anna Tillery*, Angela Aherrera*, Rui Chen, Joyce Lin, Mina Tehrani, Donia Moustafa, Jana N. Mihalic, Andrew Schultze, Ana Navas-Acien, Ana M. Rule   Johns Hopkins Bloomberg School of Public Health, Baltimore MD  

Introduction: Electronic cigarettes have rapidly evolved from “mod” devices mostly used by smokers as a risk reduction strategy with adjustable settings, to “pod” devices with disposable cartridges, marketed to youth never smokers. Inhalation of metals that have been found in e-cigarettes can lead to adverse health outcomes such as asthma and lung cancer. Evaluating e-cigarette use behaviors, device type, settings, and biomarkers are essential to understanding potential heavy metal exposures attributed to e-cigarette use.  

Objective: Assess heavy metal concentrations in blood, urine, and exhaled breath condensate (EBC) of e-cigarette users and controls by use behaviors and device type.

Methods: We recruited 17 Mod, 24 Pod, 10 smokers, 14 dual users (cigarette and e-cigarette), and 30 non-users. Sociodemographic characteristics, e-cigarette/tobacco use behaviors, and device characteristics were collected by survey. Blood, urine, EBC and aerosol samples were analyzed for heavy metals using ICP-MS. Data was corrected for background and limit of detection. Chi-squared tests for categorical variables, ANOVA tests for continuous variables, and linear regressions were used to assess relationships between variables and user groups.

Preliminary Results: Never smokers and younger users were more likely to use higher nicotine concentration than former smokers. Significant differences (p<0.05) were found between user group and Cd, Mn, Ni, and Zn in blood. Cr and Ni in mod users was 1.2 and 1.5 times higher, respectively, than pod and nonusers.  

Conclusions: Differences in user behaviors and device type determine exposure to certain metals. Blood Cd, Mn, Ni, and Zn were associated with user group.  

This research was supported by a grant from NIEHS/FDA (R01 ES030025); AT is supported by a grant from the U.S. Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health to the Johns Hopkins Education and Research Center for Occupational Safety and Health (T42 OH0008428) and AA is supported by a training grant from NHLBI (T32HL072748)  

Wilmsen, Kai

COPD Alters Bronchial Epithelial Wound Closure Dynamics In vitro

Chronic Obstructive Pulmonary Disease (COPD), a progressive and irreversible disease of the lungs, can be characterized in vitro via cultured patient-derived bronchial epithelial cells. Patient-derived COPD cells exhibit reduced barrier function, partial EMT (pEMT), reduced ciliation and ciliary function, and the induction of collective cellular motion within the epithelial sheet. Interestingly, the collective motion of COPD cells in vitro occurs under normal culture conditions in intact and undisturbed epithelial cells. To better understand the functional consequences of increased collective motion in COPD we are striving to understand the cellular processes that are driving the observed motion and if these changes impact migration dependent processes like wound closure. Scratch wound closure assays of Normal and COPD patient-derived cells cultured at air-liquid interface (ALI) uncovered interesting differences in the cellular dynamics of wound closure. 15-hour timelapse imaging of wound closure showed COPD cells exhibit slightly decreased wound closure compared to Normal cells, however, measurements of cell movement speed were unchanged between the two groups. Additionally, normal cells have their movement concentrated at the leading edge of the wound, with cellular velocity tapering off at further distances in sharp contrast with COPD cells which move at the same speed regardless of proximity to the wound edge. Along with these findings, we have observed altered localization of the focal adhesion protein vinculin and the planar cell polarity proteins vangl1 and frizzled 6 in COPD cells, suggesting that the collective motion observed in COPD cells may be due to cellular reprogramming that primes COPD cells for migratory processes even in the absence of a wound edge. Further research is required to understand the gene and/or protein level changes leading to the observed COPD phenotype, and to better understand if COPD cells can resolve a wound and return to their steady state condition.

Engineering Abstracts

Feng, Leyang

US Methane Emission Anomaly Detection Using a Novel Geostatistical Inverse Modeling Algorithm

Leyang Feng1, Julianne Chung2, Jiahua Jiang3, Arvind K. Saibaba4, Lee T. Murray5, Xueying Yu6, Scot M. Miller1

1. Department of Environmental Health and Engineering, Johns Hopkins University, Baltimore, MD
2. Department of Mathematics, Emory University, Atlanta, GA
3. School of Mathematics, University of Birmingham, UK
4. Department of Mathematics, North Carolina State University, Raleigh, NC
5. Department of Earth and Environmental Sciences, University of Rochester, Rochester, NY
6. School of Sustainability, Stanford University, Stanford, CA

Existing atmospheric methane observations have yielded enormous insight into methane emissions across the United States, particularly for the oil and natural gas (O&G) sector. These observations, including measurements from surface towers, aircraft campaigns, and satellite platforms, have been used in numerous studies to quantify methane emissions from regional scale to continental scale. However, fewer studies discuss how heterogeneous observations from multiple data sources can reveal methane ‘supper-emitters’ caused by abnormal process conditions and accidental leakages in the O&G industry. In this work, we conduct a series of Observing System Simulation Experiments (OSSEs) to detect methane emission anomalies across the U.S., using a novel geostatistical inverse modeling algorithm sdHyBr coupled with an atmospheric transport model GEOS-Chem and methane observations from the TROPOMI satellite sensor. We design multiple emissions and observation scenarios to test and evaluate the emission flux reconstruction capability of the sdHyBr algorithm. The results help us to better understand the US methane emission in two aspects: detecting hotspots of large methane emissions and revealing different spatiotemporal patterns in different oil and gas production basins of methane emissions.

Gaeta, Dylan

Constraining U.S. emissions of sulfuryl fluoride with atmospheric measurements and a geostatistical inverse model 

Dylan C. Gaeta, Jens Mühle, Isaac J. Vimont, Lei Hu, Molly Crotwell, Mingyang Zhang, Kathryn McKain, Bianca C. Baier, John B. Miller, Benjamin R. Miller, Jianing Bao, Scot M. Miller  

Sulfuryl fluoride (SO2F2) has been accumulating in the global atmosphere since the 1970s, with ambient air concentrations exceeding 2.5 parts per trillion (ppt) today. SO2F2 is a synthetic pesticide used primarily for structural fumigation, and increasingly for agricultural and commodity fumigation. It is also a potent greenhouse gas (GHG) with an atmospheric lifetime of 40 years. Worldwide use of SO2F2 has surged since the use of methyl bromide (CH3Br) was largely phased out under the Montreal Protocol. Recent work from the Advanced Global Atmospheric Gases Experiment (AGAGE) indicates that global emissions of SO2F2 have recently reached a historic high of 3.0 Gg SO2F2 yr-1. However, under the current UNFCCC emissions reporting guidelines, countries are not required to report their emissions of SO2F2, and thus SO2F2 is not included in most national GHG emissions inventories, leading to a scarcity of information on the global distribution and magnitude of SO2F2 emissions. To fill this information gap, we use flask-sample measurements of SO2F2 from the NOAA Global Greenhouse Gas Reference Network (GGGRN) and a geostatistical inverse model (GIM) to provide a constraint on the magnitude and spatial distribution of SO2F2 emissions across the continental U.S. We find that California emits the vast majority (>90%) of U.S. SO2F2 emissions, with the largest emissions coming from Southern California (Los Angeles, Orange, and San Diego Counties) – where drywood termite infestations of wooden structures are common. Outside of California, SO2F2 emissions are sparse and rarely detected by the NOAA GGGRN. We will present insights gained from the NOAA GGGRN measurements and our inverse modeling work, including a case study on California’s SO2F2 emissions and reconciliation with California state SO2F2 usage records. Our results suggest that 2/3rd of SO2F2 used is absorbed or destroyed during fumigation, which has implications for human exposure to SO2F2 and SO2F2 residues.   

Lai, Junxi

Combining Capillary Electrophoresis and Iron Colorimetry: A New Analytical Approach Applied to the Monitoring and Management of Growth Media for Mammalian Cells

Background:  Mammalian cell growth media supports essential processes in modern biomanufacturing. The trend in biomanufacturing is towards chemically defined media (CDM, i.e., medium free of added proteins or other cell-derived ingredients that pose risks of bacterial or viral contamination). Components often found in other biofluids supporting certain essential biological processes are lacking in CDM. Without proteins dedicated to Fe transfer and storage (e.g., transferrin and ferritin), Fe is free to interact with all the other components in CDM, thus it needs more attention and better management. Currently available analytical methods can’t discern FeII from FeIII well enough in CDM.

Objective/aim:  Development of an analytical method to determine FeII and FeIII concentrations in CDM aliquots.  

Methods:  Following the evaluation pathway of traditional colorimetric methods (e.g., ferrozine method), we propose to add a pair of Fe selective chelating agents (BPDC and HBED) simultaneously to CDM aliquots so that Fe oxidation state can be stabilized. All the FeII species should be converted to FeII-BPDC complex and all the FeIII species should be converted to FeIII-HBED complex at equilibrium. A separation technique, capillary electrophoresis (CE), will be employed to assist the quantification of FeII-BPDC and FeIII-HBED complexes, so that interference from constitute CDM components can be minimized.  

Results:  Details of the proposed method regarding dilution rate and pH adjustment of CDM aliquot as pretreatment, concentrations of BPDC and HBED to add, and CE method quantification of FeII-BPDC and FeIII-HBED complexes have been decided based on results from preliminary results. Our method reported correct concentration of FeII and approximately 20% lower concentration of FeIII when it’s applied to CDM samples with known amount of FeII and FeIII prepared in our laboratory.

Conclusion:  Our method can report accurate concentrations of FeII in CDM aliquots. It needs further optimization to report accurate concentrations of FeIII

Litwin, David

DupuitLEM and the Search for Fundamental Insights into the Coevolution of Landscape Hydrology and Geomorphology

Authors: David G. Litwin (1), Gregory E. Tucker (2,3), Katherine R. Barnhart (4), Ciaran J. Harman (1,5) 1. Department of Environmental Health and Engineering, Johns Hopkins University, Baltimore, MD, USA 2. Cooperative Institute for Research in Environmental Sciences (CIRES), University of Colorado, Boulder, CO, USA 3. Department of Geological Sciences, University of Colorado, Boulder, CO, USA 4. U.S. Geological Survey, Geologic Hazards Science Center, Golden, CO, USA 5. Department of Earth and Planetary Science, Johns Hopkins University, Baltimore, MD, USA

Most long-term landscape evolution models still operate under the assumption that discharge scales with upslope accumulated area. While this can be appropriate for answering certain kinds of questions, it overlooks a wealth of hydrological knowledge and prevents us from asking hydrological questions of landscape evolution models. For instance, how do subsurface properties and climate impact drainage density? Where and why do variable source areas develop? The hydrological model needed to answer these questions must be sophisticated enough to capture evapotranspiration, subsurface lateral flow, and the development of saturated areas, while being efficient enough to evolve the landscape over thousand-to-million-year timescales. However, this does not necessarily mean a new model should be built from scratch, which would be time consuming and duplicate many existing high-quality tools. To answer these questions, we took a community modeling approach, building a groundwater component for Landlab, and integrating other process components to develop a coupled modelling platform called DupuitLEM. While model process components like those in Landlab can be simply coupled in a script, this is often just the beginning of what is needed to manage the full workflow of model testing, handling parameter sets, sequencing component execution, managing outputs, and post processing. Here we will present highlights of the research that has been accomplished with this model, along with thoughts on model development that may be useful to others interested in using community software to jumpstart investigations of processes across broad ranges of timescales.  

Ortiz, John

Martial Atmospheric Pressure Fluctuations Can Vent Underground from Habitable Depths

Background: The existence of methane on Mars is a topic of significant interest because of its potential association with subsurface microbial life. Measurements of methane in the atmosphere of Mars indicate that its abundance fluctuates over time. Although the source of methane is unknown, it most likely comes from below the surface of Mars. If methane is currently being produced by living microbes, it would have to be at depths of at least 200 m in order to support life. This begs the question of how to produce short-term atmospheric methane variations if the source is relatively deep underground.

Objective/aim: I am to demonstrate that atmospheric pressure fluctuations (i.e., “barometric pumping”) on Mars are capable of delivering methane to the surface from habitable depths rapidly, and could be a way to explain the observed short-term atmospheric concentration variations.

Methods: I performed numerical flow and transport simulations in fractured rock using Mars atmospheric pressures as surface forcing.  

Results: My flow and transport simulations in fractured rock demonstrate that barometric pumping is capable of producing significant surface fluxes of underground methane. The simulations confirm the clear connection between atmospheric pressures and surface methane flux hypothesized in previous works. Driven by atmospheric pressure fluctuations acquired by the Curiosity barometric record, the overall simulated surface seepage pattern at Gale crater is highly seasonal. It is noteworthy that the seasonality of fluxes generated in our model is reasonably consistent with that of atmospheric methane abundance variations found in previous research.  

Conclusion: Because barometric-pressure pumping is able to vent subsurface methane from potentially habitable depths, this work reinvigorates the hypothesis that methane on Mars could be generated by microbial life. 

Wallace, Veronica

Contributions of non-antibiotic pharmaceuticals to antimicrobial resistance

Authors: Veronica J. Wallace1, Eric G. Sakowski2, Sarah P. Preheim1, Carsten Prasse1 Johns Hopkins University1, Mount St. Mary's University2

Background: Substantial evidence points to overuse of antibiotics and presence of antibiotics in the aquatic environment as key aspects contributing to the rise in antibiotic resistance throughout the world. However, there is increasing evidence that other anthropogenic compounds in the environment including pesticides and non-antibiotic drugs can have antibacterial effects and may also play a role in the development and proliferation of antimicrobial resistance.

Objective/aim:  As a proof-of-principle that various non-antibiotic compounds may contribute to antimicrobial resistance in humans or the environment, we focused on selected antiviral and antidepressant drugs that have been shown to have antibacterial properties and have been detected in wastewater, surface water, and ground water. We aimed to determine whether exposure to non-antibiotic drugs with antibacterial properties would alter the phenotypic and genotypic resistance profiles of gram-negative E. coli.

Methods:  Resistance was first demonstrated by isolating wild type E. coli that had been passaged in the presence of non-antibiotic drugs and challenged with higher concentration drug exposures. The resistant isolates were then challenged in vitro with common antibiotics. We conducted whole genome sequencing of resistant isolates to identify mutations conferring antibiotic cross-resistance and infer mechanisms of non-antibiotic drug resistance.  

Results: Our results show E. coli could easily develop resistance to antiviral and antidepressant drugs upon repeated exposure. Most importantly, resistance to selected common antibiotics including amoxicillin, erythromycin and clarithromycin, tetracyline and trimethoprim, was observed in antiviral- and antidepressant-resistant E. coli. Whole genome sequencing identified both known antimicrobial resistance mutations and novel mutations likely conferring broad cross-resistance to antibiotics.  

Conclusion: Our results highlight that exposure of E. coli to antiviral and antidepressant drugs can result in antibiotic cross-resistance. Some antibiotic and non-antibiotic drugs may act on similar pathways, highlighting the importance of studying environmental mixtures, synergistic antimicrobial effects, and the risk of developing antibiotic cross-resistance due to non-antibiotic drug exposures.  

Wang, Shen

Linking Energy Sector and Air Quality Models Through Downscaling: Long Run EGU Siting and Dispatch to Account for Spatial and Temporal Variability

Modeling the air pollution implications of long-term energy transitions requires a downscaling process as an intermediate step between national-scale energy models and fine-scaled air quality models. Traditional “Grow-in-Place” (GIP) downscaling methods assume that future patterns of generator siting and operation will be similar to those in the past. However, rapid technological change and shifting policy might yield very different future spatiotemporal patterns of power emissions. Here, we propose a “Site-and-Grow” (SAG) and “Temporal Dispatch Model (TDM)” downscaling framework to couple the Electricity Market Module (EMM) of the National Energy Modeling System (NEMS) with the Community Multi-scale Air Quality (CMAQ) model to simulate future changes in emissions from power sector. The SAG focus on spatial disaggregation which consists of two steps - a modified generation expansion model to downscale regional energy information and GIS-based screening and siting to specify the final county-level placement. The TDM targets on temporal allocation, which is power system dispatch model taking the SAG projected generation mix and a multi-variable linear regression model to projected future hourly load demand and renewable profiles as inputs. It operates projected future power system and outputs power emissions reflecting impacts of simulated meteorological conditions. The method is implemented in one EMM region (Carolinas and Virginia) as a case study. We compare GIP and SAG-TDM methods downscaled among four NEMS scenarios (base case, high natural gas consumption, high penetration of electric vehicles, and marine vessel electrification in ports). The results indicate the bias exists in GIP downscaled emissions in terms of difference in spatial, temporal and projections compared to SAG-TDM downscaled emissions which accounts for more realistic representation of fundamental changes in system, economics, policy and climate.  

Wilcox, Stephanie

Modeling the Long-Term Impacts of Municipal Ordinance Driven Building Decarbonization

Background: In the U.S. in 2021, commercial and residential energy consumption represented 35% of total national greenhouse gases (GHG) emissions. However, the decentralized and long-lived nature of space conditioning and electricity investment decisions limits the rate of GHG reductions in the two sectors. To incentivize an energy transition from fossil-based to low-carbon sources in buildings, several cities have developed GHG-limiting ordinances that restrict the installation of natural gas pipelines and/or the combustion of substances emitting GHGs above a specified rate. Existing literature indicates that GHG-limiting ordinances directly impact infrastructure investments, electricity and fuel prices, and energy access; but they only assess electrification, do not optimize intermediate energy transition steps, and omit modeling construction starts. This study explores these gaps with a case study on New York City’s (NYC) GHG-limiting ordinance, Article 2317-A.

Objective: How do GHG-limiting ordinances applied to new residential and commercial construction starts impact NYC’s energy system? In particular, how do energy sources, emissions, and system and consumer costs change?

Methods: To address the three gaps identified above, this study uses a combined approach of econometrics-based construction start modeling and cost-optimal energy planning via linear programming. First, the work introduces a two-step approach to predicting floor area added from 2023 to 2040 by building category. The method implements a linear regression model to predict building starts followed by distribution fitting and Monte Carlo simulation with Latin Hypercube sampling. Second, the work extends the U.S. EPA’s City-Based Optimization Model for Energy Technologies to include natural gas upgrade decisions and pipeline operational constraints (Kaplan and Isik 2020). 

Results/Conclusion: The predicted construction starts indicate that the GHG-limiting ordinance impacts NYC boroughs unequally and will most affect the residential sector. Analysis shows that to achieve long-term energy goals, NYC likely must pass policies targeting existing buildings in addition to the GHG-limiting ordinance.  

Xinxiu, Tian

Predicting Floor-Based Variability in Composition and Concentration of Indoor Aerosol of Outdoor Origin in a High-Rise Multi-Family Building 

Xinxiu Tiana, Bryan E. Cummingsb, Michael S. Waringb, Marianne F. Touchiec,d, Peter F. DeCarloaaDepartment of Environmental Health and Engineering, Johns Hopkins University, USA bDepartment of Civil, Architectural, and Environmental Engineering, Drexel University, USA cDepartment of Mechanical and Industrial Engineering, University of Toronto, Canada dDepartment of Civil and Mineral Engineering, University of Toronto, Canada


In this work, we model how the differences in building ventilation rates by floor and variation in outdoor aerosol concentration and composition impact indoor aerosol within a high-rise multifamily building. In the absence of strong indoor sources, outdoor-originated aerosols are the main contributor to indoor aerosols. Upon transport to the indoor environment, both concentrations and chemical compositions of outdoor aerosols are modified. Many studies show the linkage of indoor concentrations of ambient PM2.5 to various factors, including infiltration, ventilation rates, deposition and filtration, and phase change of semi-volatile material during outdoor-to-indoor transport. We demonstrate that inconsistent airflow patterns caused by natural and mechanical ventilation within a single building can induce differences in indoor concentrations of outdoor PM2.5. Airflow and pollutant simulations were performed with a CONTAM building model to obtain the indoor-outdoor ratio of a non-volatile, non-reactive test species under different ambient temperatures. Chemical compositions of ambient PM2.5 were reconstructed from regulatory monitoring data based on modified PM2.5 mass reconstruction techniques. Outdoor PM concentration and composition were used to calculate indoor ratios of semi-volatile species including NO3, HOA, and OOA using linear relationships developed by previous modeling work. Indoor-outdoor ratios and, by extension, concentrations and composition of particulate chemical species showed variation across seasons and by floor due to non-uniform building ventilation air distribution. This work demonstrates that there are significant temperature-based and floor-based variations in indoor-outdoor ratios of chemical species within a single building, with variations in chemical compositions of ambient PM, resulting in strong seasonal and floor-based variations in indoor concentrations and compositions of ambient species.

Xu Fei, Esther

Dynamic uncertainty quantification of catchment transit time and StorAge Selection distributions using an adaptive non-parametric Bayesian framework  

Interrogating tracers using numerical models is a powerful approach to track water fluxes and understand catchment functions. StorAge Selection (SAS) functions implemented in a dynamic mass-conservative transport model capture the preferential sampling of streamflow to storage and the resulting transit time distributions. However SAS approaches typically require the functional form of the SAS function to be specified a-priori. MESAS (Multi-scale Estimation of StorAge Selection), is a non-parametric self-adaptive model that learns the shape of the SAS function from stable water isotope data along with other flux observations without requiring a-priori statistical distribution. While MESAS aims to extract the maximum possible information about catchment structure from the data, the estimated distributions are uncertain. This uncertainty originates from 1. ontological uncertainty about controls on the time-variability of the SAS function, 2. the finite length and resolution of input time series data, and 3. measurement error in that data. These errors propagate through transport process and accumulates in system. We propose to use a two-step uncertainty quantification method to further explore the estimation of SAS function: 1. estimate uncertainty introduced by temporal downscaling of the isotopic composition of precipitation, and 2. estimate uncertainty bounds on the SAS function using a Bayesian approach, i.e., particle Markov chain Monte Carlo (PMCMC).

We used our approach to quantify uncertainties in time-varying storage mobilization and transit times at a field site, Pond Branch watershed in MD. This is forested watershed with a unique bimodal hydrologic response to events, which indicates the existence of its diverse flow paths. A 5-year record of weekly precipitation stable water isotope, 12-hourly streamflow stable water isotope, and other inflow and outflow time series has been collected here. This research advances our ability to reveal the connection between passive tracer samples and runoff generation, and ultimately the runoff generation mechanism in similar watershed. 

Zhang, Mingyang

Trends in U.S. ethane emissions and implications for fugitive leaks from the oil and gas industries

Oil and natural gas (O&G) production activities have changed markedly across the U.S. over the past several years. However, the impacts of these changes on fugitive oil and gas emissions are not clear. For example, between 2017 and 2020, natural gas production grew at a rate 2.7 times faster than the previous decade. However, the onset of the COVID-19 pandemic dramatically impacted the world economy, including the O&G sector. The shutdown impact on the U.S. O&G sector is multifold, including disrupted production and processing and possibly delayed maintenance.

In this study, we examine US ethane (C2H6 ) emissions from 2015 onward as a lens to understand changes in O&G industry fugitive emissions. Ethane is a non-methane hydrocarbon that is predominantly emitted from O&G operations in the U.S., and ethane can therefore act as an important tracer for fugitive O&G emissions. We specifically use ethane observations from towers and aircraft, collected by the NOAA Global Monitoring Laboratory and partner organizations, for years 2015 through 2020. We further pair these observations with a geostatistical inverse model to quantify whether U.S. O&G ethane emissions have changed and by how much. Overall, we find that U.S. ethane emissions increased between 2015-2020 but that the increase is much less than reported changes in U.S. ethane production. We will also discuss changes during the pandemic and the sensitivity of estimated trends to possible changes in atmospheric chemistry. 

Zhang, Zhuoyue

Elucidation of reaction pathways of ring cleavage products from chlorination of 13C labeled phenols

Background:  For more than a century, chlorine has been widely applied in water treatment as disinfectant for inactivation of pathogens in drinking water. However, its reactions with organic substances can lead to the formation of a variety of toxic DBPs. Among the known DBP precursors, phenols have been shown to be particularly important due to their widespread occurrence in both natural organic matter and anthropogenic compounds and their high reactivity with chlorine. Despite decades of research on the formation of toxic disinfection by-products (DBPs) in the reaction of free chlorine with phenols, uncertainties remain regarding the identity of ring cleavage products and the reaction pathways involved.

Objective/aim:  To systematically elucidate the reaction pathways leading to the formation of the C4- and C2-DBPs during chlorination of phenols.

Methods:  We synthesized a series of stable-isotope-labeled ethyl parabens (EP) containing 13C labels at different positions of the molecule and investigated the formation of DBPs via a novel amino acid reactivity assay and liquid chromatography-high resolution mass spectrometry (LC-HRMS), which can directly detect the 13C isotopic differences in the DBPs.

Results:  We found that, besides the known C2-DBPs chloroacetic acids, α, β-unsaturated C4-dialdehydes and C4-dicarboxylic acids were formed as previously unidentified ring cleavage products. The formation of C4- and C2-DBPs inheriting 13C labels from corresponding EPs revealed the existence of four general ring cleavage pathways. Moreover, quantification of the different 13C labeled DBPs enabled further assessment of the contribution of different pathways.

Conclusion:  Ring cleavage pathways in the reactions of free chlorine with phenols are more complex than previously reported. These novel DBPs and different pathways should be considered in the reaction of free chlorine with phenols in all contexts.  

Zhong, Shengbang

Modeling the Impacts of Climate Change on Weathering Reactions and Water Quality in Mountain Watersheds

Climate change increasingly influences water resources in mountainous regions through earlier spring snowmelt and runoff, declining snowpack volume and reduced the duration and intensity of snowfall. Several observational studies in Colorado River Basin document changes in both hydrology and water quality due to climate change. The specific objective of this research is to develop reactive transport models that can reproduce the multi-decadal trends in water quality observed in some mountain watersheds in the Upper Colorado River Basin. My overall goal is to advance understanding of hillslope scale pyrite weathering process and predict the influence of further decreasing water tables on weathering rates, solute export, and water quality, resulting from climate change. To achieve my goal, I develop hydrologic and multicomponent reactive transport models, which are based on saturated-unsaturated flow equations and multi-component reactive transport equations. The numerical model is conducted using PFLOTRAN, a massively parallel reactive flow and transport model for describing subsurface processes. The modeling shows water table variations throughout the year and stream ion concentrations, which match the results from site observations. Also, it suggests that a falling water table could produce an increase in metals and SO4. For future projections, I will seek collaborations with research groups that have made future projections of snowpack dynamics, snowmelt and recharge, for example at NCAR (National Center for Atmospheric Research) to obtain hydrologic forcing data. Alternatively, I will also consider hypothetical scenarios to improve understanding of the influence of water table decline on pyrite oxidation rates. Specific questions include: (i) if water table decline continues steadily, what future trends can we expect in stream water chemistry?  (ii) Is there an upper limit to pyrite oxidation rates, for instance due to inability of oxygen to penetrate to greater depths as the water table falls significantly?