@article{mccord_parsons_bittner_jumbe_kabwe_pedit_serenje_grieshop_jagger_2024, title={Carbon Monoxide Exposure and Risk of Cognitive Impairment Among Cooks in Africa}, volume={2024}, ISSN={["1600-0668"]}, DOI={10.1155/2024/7363613}, abstractNote={We use air pollution exposure measurements and household survey data from four studies conducted across three countries in sub‐Saharan Africa (SSA) to analyze the association between carbon monoxide (CO) exposure from cooking with biomass and indicators of cognitive impairment. While there is strong evidence on the relationship between ambient air pollution exposure and cognitive impairment from studies in high‐income countries, relatively little research has focused on household air pollution (HAP) in low‐income country settings where risks of HAP exposure are high. This study is the first to our knowledge to focus on the association between HAP exposure (specifically CO exposure) and cognitive impairment across diverse settings in SSA. We use 24‐hour measurements of primary cooks’ exposure to CO across four study sites: urban Zambia ( n = 493); urban Malawi ( n = 130); rural Malawi ( n = 102); and urban Rwanda ( n = 2,576). We model the estimated percent carboxyhemoglobin (%COHb) of cooks and map values to a toxicological profile for risk of cognitive impairment. We find that across all study settings, cooks’ average %COHb levels are below levels of daily concern, but that cooks who use charcoal for preparing greater than 40% of meals are more likely to spend additional time at higher levels of risk. For the urban Zambia sample, we compare %COHb and frequency of charcoal use to a series of cognitive test scores and find no consistent relationships between %COHb and cognitive test scores. High levels of daily CO exposure from cooks across SSA highlight the potential for longer‐term negative cognitive (and other) health outcomes motivating additional research and efforts to characterize and mitigate risk.}, journal={INDOOR AIR}, author={McCord, Ryan and Parsons, Stephanie and Bittner, Ashley S. and Jumbe, Charles B. L. and Kabwe, Gillian and Pedit, Joseph and Serenje, Nancy and Grieshop, Andrew P. and Jagger, Pamela}, year={2024}, month={Jun} }
 @misc{weyant_amoah_bittner_pedit_codjoe_jagger_2022, title={Occupational Exposure and Health in the Informal Sector: Fish Smoking in Coastal Ghana}, volume={130}, ISSN={["1552-9924"]}, DOI={10.1289/EHP9873}, abstractNote={Vol. 130, No. 1 Research LetterOpen AccessOccupational Exposure and Health in the Informal Sector: Fish Smoking in Coastal Ghana Cheryl L. Weyant, Antwi-Boasiako Amoah, Ashley Bittner, Joe Pedit, Samuel Nii Ardey Codjoe, and Pamela Jagger Cheryl L. Weyant https://orcid.org/0000-0002-0654-6456 School for Environment and Sustainability, University of Michigan, Ann Arbor, Michigan, USA Search for more papers by this author , Antwi-Boasiako Amoah Centre for Climate Change and Sustainability Studies, University of Ghana, Legon, Ghana Environmental Protection Agency (Ghana), Accra, Ghana Search for more papers by this author , Ashley Bittner https://orcid.org/0000-0003-0402-6702 Department of Civil, Construction, and Environmental Engineering, North Carolina State University, Raleigh, North Carolina, USA Search for more papers by this author , Joe Pedit University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA Search for more papers by this author , Samuel Nii Ardey Codjoe https://orcid.org/0000-0002-6567-0262 Regional Institute for Population Studies, University of Ghana, Legon, Ghana Search for more papers by this author , and Pamela Jagger Address correspondence to Pamela Jagger, School for Environment and Sustainability, University of Michigan, 440 Church St., Ann Arbor, MI 48109 USA. Email: E-mail Address: [email protected] School for Environment and Sustainability, University of Michigan, Ann Arbor, Michigan, USA Search for more papers by this author Published:12 January 2022CID: 017701https://doi.org/10.1289/EHP9873AboutSectionsPDF ToolsDownload CitationsTrack Citations ShareShare onFacebookTwitterLinked InReddit IntroductionFish preservation by smoking is an occupation where workers (primarily women) are exposed to gases and particles from wood smoke for more than 5 h daily (Adeyeye and Oyewole 2016). The scale of this issue is large; we estimate that there are 6 million fish smokers on the West African coast (uncertainty: 0.8–10 million).MethodsWe used a cross-sectional design to explore differences in exposure and self-reported health symptoms between women engaged in occupational fish smoking (fish smokers, N=308) and those in other occupations, including business- and tradeswomen, fish salters, tailors, hairdressers, and others (controls, N=152). Households were randomly selected from two small coastal cities in Ghana (Moree and Elmina). Fish smoker and control households were geographically well mixed at an ∼2:1 ratio.A structured survey was used to collect sociodemographic and health symptom data from each household’s highest income earning woman. The health questionnaire was modeled on the Ghana Demographic and Health Survey and included 1-y and 2-wk symptom recall (Table 1).Table 1 Symptom prevalence in controls (C) and fish smokers (FS).Table 1 lists symptom prevalence in controls (C) and fish smokers (F S) for 1-y and 2-week recall periods. Each symptom’s association with fish smoking was tested using chi-squared tests (lowercase p-values shown) and logit models (odds ratio, 95 per cent confidence interval, and lowercase p-values shown). The association between each symptom and C O exposure (24-hour and 8-hour) and particulate matter begin subscript 2.5 end subscript were assessed with lowercase t-tests (lowercase p-values shown).SymptomsSymptom recall in past year and symptom relationship withSymptom recall in past 2-wk and symptom relationship withFish smokingCOPM2.5Fish smokingCOPM2.524ha8ha24h24ha8ha24hC (%)FS (%)χ2 (p-value)Logit [OR (95% CI); p-value]bt-test (p-value)t-test (p-value)t-test (p-value)C (%)FS (%)χ2 (p-value)Logit [OR (95% CI); p-value]bt-test (p-value)t-test (p-value)t-test (p-value)Eye symptoms Poor visionc———————16484×10–116.8 (3.6, 12.9); 2×10–90.03*0.030.15 Very impaired visionc———————160.049.5 (1.4, 65.5); 0.020.930.75— Burning eyes60881×10–127.8 (4.0, 15.2); 1×10–90.010.03*0.2848802×10–126.2 (3.5, 11.0); 5×10–100.180.100.28 Sticky eyes13180.152.1 (0.9, 4.7); 0.080.940.750.848130.091.9 (0.7, 4.9); 0.180.450.280.81Neurological symptoms Headache89920.331.7 (0.7, 4.0); 0.240.150.120.3166760.022.2 (1.2, 3.8); 0.010.050.030.64 Difficulty concentrating60660.231.2 (0.7, 1.9); 0.570.200.210.0745540.081.4 (0.8, 2.3); 0.200.200.190.77 Dizziness45580.0092.8 (1.6, 4.8); 2×10–40.030.030.6826360.031.9 (1.1, 3.4); 0.020.220.250.43 Forgetfulness71730.510.9 (0.5, 1.7); 0.840.560.780.7555640.071.1 (0.7, 1.9); 0.630.150.150.57 Approx. neurological impactd28400.022.4 (1.4, 4.3); 3×10–30.250.220.1114180.341.5 (0.7, 3.0); 0.300.430.450.71Skin issues Burns35460.021.4 (0.9, 2.4); 0.160.020.010.038190.0012.9 (1.2, 6.6); 0.010.060.050.45 Burns from cookinge3465×10–110.11 (0.05, 0.3); 6×10–73×10–42×10–5———————— Moderate and severe burnse18160.470.6 (0.3, 1.2); 0.180.170.110.01——————— Rashes10150.171.4 (0.7, 3.0); 0.380.290.250.337110.161.9 (0.7, 4.9); 0.200.960.910.33 Skin irritationf49580.082.3 (1.4, 3.8); 2×10–30.400.390.1230350.271.8 (1.0, 3.1); 0.040.780.630.16Respiratory infection symptoms Fatigueg97990.3019.3 (0.7, 501); 0.080.060.10—86910.111.3 (0.6, 2.8); 0.520.220.130.68 Vomitingg45410.461.1 (0.6, 1.8); 0.780.070.040.2925230.611.0 (0.6, 1.8); 0.960.970.770.57 Feverg78800.721.4 (0.8, 2.8); 0.270.510.630.1246490.571.2 (0.7, 1.9); 0.590.900.850.42 Chest infection67790.0062.0 (1.1, 3.5); 0.020.220.150.6453630.031.8 (1.1, 3.0); 0.030.660.790.91 Approx. respiratory infectionh58680.041.7 (1.0, 2.9); 0.060.640.450.6230340.411.4 (0.8, 2.4); 0.210.430.660.97 Approx. resp. infection with coughi27370.022.1 (1.2, 3.7); 0.010.530.360.079130.172.2 (1.0, 5.1); 0.060.990.150.90Respiratory distress symptoms Shortness of breath53570.552.4 (1.3, 4.3); 0.010.080.090.4843430.901.7 (0.9, 3.0); 0.080.550.480.47 Difficulty breathing/chest tightness32450.0071.9 (1.1, 3.1); 0.020.650.720.7528350.131.5 (0.9, 2.6); 0.130.450.590.37 Wheezing or whistling in chestj12220.0072.8 (1.3, 5.9); 0.010.710.680.38——————— Chronic cough53560.421.3 (0.8, 2.2); 0.320.980.830.0433360.311.3 (0.7, 2.2); 0.400.040.060.32 Cough with blood120.34————020.24———— Produced phlegm57700.0032.4 (1.4, 4.3); 2×10–30.570.460.1440520.021.8 (1.1, 3.0); 0.030.880.900.29 Approx. COPDk150.0813.4 (0.9, 196); 0.060.860.900.71130.382.1 (0.1, 53.6); 0.670.870.76— Approx. asthmal120.40————110.96———— N1523084601511512615230846015115126Note: Each symptom’s association with fish smoking was tested using chi squared tests and logit models for 1-y and 2-wk recall periods (odds ratios and 95% confidence intervals shown). The association between each symptom and CO exposure (24- and 8-h) and PM2.5 were assessed with t-tests. —, statistical test could not be conducted; Approx., approximated; CI, confidence interval; CO, carbon monoxide; COPD, chronic obstructive pulmonary disease; OR, odds ratio; PM2.5, fine particulate matter (PM ≤2.5μm in aerodynamic diameter).aSymptom vs. CO exposure t-tests were conducted with log-transformed CO. * indicates p<0.05 on a non log-transformed scale.bControl variables included household characteristics (i.e., size, town, number of children), wealth indicators (i.e., assets and home quality), personal characteristics of the woman (i.e., age, marital status, education), and environmental risk indicators (household cook, cigarette smokers in the home, charcoal cooking, kerosene lighting, use of an improved cookstove, number of fish smoking ovens owned, number of unowned ovens in a 20-m radius, and if they lived/worked within 20m of trash burning).cThe vision question was asked in the present tense and did not have a recall period.dApproximated neurological impacts are cases with all of headache, difficulty concentrating, dizziness, and forgetfulness.eSources of burns and severity were only asked for the 1-y recall period.fSkin irritation effect likely from fish handling.gSymptoms may have causes other than respiratory diseases.hRespiratory infection was approximated as present if all of fatigue, fever, and chest infection were experienced.iRespiratory infection with cough are considered present if all symptoms of respiratory infection occurred as well and cough and phlegm.jWheezing symptoms were not asked for 2-wk recall.kApproximated COPD are cases with chronic cough, phlegm, shortness of breath, difficulty breathing, wheezing, and are >40 years of age.lApproximated asthma are cases with chronic cough and wheezing, without produced phlegm or coughing up blood.Twenty-four–hour carbon monoxide (CO) and fine particulate matter (PM2.5) concentrations were collected near the breathing zone from a subset of participants (CO: N=151, PM2.5: N=26). CO sensors (Lascar Electronics, Model EL-USB-CO) were calibrated before and after fieldwork (198±4 ppm CO in standard air), and data synthesized into 24-h and sub–24-h averages (maximum 8-h, 1-h, and 15-min). PM2.5 samples were collected on 2-μm polytetrafluoroethylene membrane filters, using a personal air sampling pump (SKC Inc., model AirChek XR5000) connected to a single-stage impactor (MSP Corporation, Model 200 Personal Environmental Monitor).The association between exposure and fish smoking was assessed using t-tests and ordinary least squares regression models, controlling for household and individual level variables, including proximity to other sources of combustion pollutants (Table 1). Relationships between health symptoms and fish smoking were tested with χ2 tests and logit models that controlled for the same variables as the ordinary least squares models. Exposure differences due to ventilation and improved smokers were tested with t-tests.Results and DiscussionPollutant ExposureCO (24-h) and PM2.5 exposures were both ∼2.6 times greater for fish smokers compared with controls (CO: 2.69, p<0.001, PM2.5: 2.61, p=0.09; Figure 1A and E). Likewise, 8-h, 1-h, and 15-min CO exposures were significantly higher for fish smokers (p<0.001, p<0.001, p=0.007, respectively; Figure 1C). The regression model shows fish smoking was a significant determinate of CO exposure [β=4.0; 95% confidence interval (CI): 0.2, 0.6; p=0.004]. PM2.5 was strongly correlated with CO (Pearson’s r=0.89). All PM2.5 exposures were greater than the annual World Health Organization safe guideline (10 μg/m3) and 92% were greater than the interim guideline (35 μg/m3; Figure 1E).Figure 1. (A) Box plot of 24-h CO exposure on a log scale. Points represent individual women’s exposures and are jittered vertically to better show each point. Occupational fish smokers were exposed to 7.7±6.9 ppm (median=5.7 ppm), and controls were exposed to 2.9±2.6 ppm (median=1.9 ppm, p<0.001). The WHO safe guidance level for CO is 6 ppm and is shown as a vertical dashed line. (B) Mean hourly CO time series for fish smokers and controls, generated using 1-min data. From 1400–2100 hours, fish smokers experienced over three times greater CO concentrations compared with controls. (C) Probability distributions of CO exposure for 24-h, 8-h, 1-h, and 15-min averages. Mean values for fish smokers and controls were 16.7±13.2 ppm and 5.7±6.0 ppm for the 8-h, 41.0±37.1 and 16.4±16.1 ppm for the 1-h, and 71.5±84.5 and 31.0±33.8 ppm for the 15-min averages, respectively. (D) CO exposure for fish smokers using open air (5.6±5.4, p=0.009 ppm), partially ventilated (sheltered, 7.2±5.5 ppm, p=9.0×10–6), and indoor smokers (12.0±10.2, p=1.1×10–6 ppm), relative to controls (2.9±2.6 ppm). (E) Box plot of 24-h PM2.5 exposure on a log scale. Fish smokers were exposed to an average of 490±430 μg/m3 and controls to 190±210 μg/m3. The WHO guidance level for PM2.5 is 10 μg/m3 (interim target, 35 μg/m3), shown as vertical dashed lines. (F) Box plot of CO exposures for women reporting good or poor vision (5.5±5.4 ppm, and 7.6±7.5 ppm). Points are distinguished as fish smokers (circles) and controls (diamonds). Note: CO, carbon monoxide; PM2.5, fine particulate matter (PM ≤2.5μm in aerodynamic diameter); WHO, World Health Organization.Exposures experienced by fish smokers were higher than from household cooking in nonfishing villages in Ghana. Fish smokers had 24-h CO exposure that was seven times greater than from household cooking in Eastern Ghana (1.1 ppm) (Lee et al. 2019); PM2.5 exposures were four times greater (129 μg/m3) (Van Vliet et al. 2019). Controls also had higher exposure than typical from cooking (2.6 times greater for CO and 1.5 times for PM2.5) (Lee et al. 2019; Van Vliet et al. 2019). A possible explanation is that fish smoking increases local ambient pollution, raising the baseline exposure of nonfish smokers.Better ventilation may reduce exposure for fish smokers (Figure 1D). Women who used open air smokers were exposed to less than half the CO compared with those using indoor smokers (p<0.01); Flintwood-Brace (2016) also found high indoor fish smoking exposures (18±13 ppm). However, even women who used open air smokers had higher exposures compared with controls (p=0.01). Counterintuitively, users of improved smokers (with chimneys) had the same level of CO exposure as users of traditional smokers (6.9 ppm and 7.0 ppm, p>0.05), and higher PM2.5 (866 μg/m3 vs. 306 μg/m3, p<0.001).Symptom PrevalenceSymptoms were more prevalent in fish smokers compared with controls (Table 1). Yet, even for controls, rates were high compared with other populations. For example, the prevalence of difficulty breathing and concentrating were 2 and 3.3 times higher, respectively, compared with biomass cooks in Malawi (Das et al. 2017).Highly Significant Associations: Eye, Neurological Symptoms, and BurnsPoor eyesight, burning eyes, and dizziness were all strongly correlated with fish smoking and these symptoms were also associated with CO exposure (Figure 1F). These symptom associations with both fish smoking and exposure are consistent with the hypothesis that occupational fish smoking can cause higher exposure to pollutants, leading to a greater health burden.Fish smokers have substantially higher rates of impaired vision that impacts daily functioning (odds ratio=6.8; 95% CI: 3.6, 12.9). The literature supports that smoke exposure may cause or exacerbate poor eyesight, including from cataracts, age-related macular degeneration, and refractive error (Hankinson et al. 1992; Seddon et al. 1996; Bourne et al. 2013; Stone et al. 2006).Fish smokers were more likely to have burns compared with controls, despite having fewer burns from cooking. Burns reported to be caused by fish smoking tended to be mild (no scars, 70%) compared with those caused by cooking (50%). Three percent of fish smokers had a severe burn in the past year, and 15% had one in their lifetime (83% were from fish smoking).For headaches, we observed associations with fish smoking and with CO exposure, but only for the 2-wk recall period. We hypothesize that because nearly all women experienced a headache over 1-y, a shorter recall would be required to discern an association. Future studies using self-reported health symptoms should aim to match recall periods with the likelihood of symptoms (e.g., fatigue should have a shorter recall, and coughing blood, longer).Respiratory SymptomsSymptoms that are indicative of severe respiratory distress of various causes were more prevalent in fish smokers compared with controls, including shortness of breath, difficulty breathing, wheezing, and cough with phlegm. Chronic cough was associated with PM2.5 exposure (p=0.04), but not with fish smoking.Respiratory symptoms associated with fish smoking did not have a relationship with CO exposure, although they are known to be associated with smoke exposure in other contexts (Van Vliet et al. 2019). We hypothesize that this is because day-to-day exposures were variable and 24-h measures only weakly represent the exposures that drive chronic health outcomes.ConclusionOccupational fish smokers experienced an elevated health burden. Rates of self-reported symptoms in fish smokers were higher than those in other occupations, most notably increased rates of poor vision. In addition, because exposure rates in controls were also high, the true health effect estimates of fish smoking relative to a clean environment may be greater than reported here. Fish smoker health may be improved by working in well-ventilated spaces and using improved smokers field tested to verify emission reductions.The health burden from fish smoking likely impacts millions of workers and is just one of many occupations that use polluting solid fuel combustion. We show here that working with wood combustion for about 5 hours per day has measurable health and exposure associations, even when used outdoors. Millions of workers in low-income countries are engaged in informal sector occupations that use solid fuel for many hours daily (e.g., brick kiln workers, charcoal producers); exposure and health measurements are needed to understand this health burden, especially in the African context.AcknowledgmentsWe are grateful to P. Morgan, Alan Feduccia Professor Emeritus of Sociology and Emeritus Director of the Carolina Population Center, University of North Carolina at Chapel Hill, for his encouragement and support on this project. Funding for this research came from the Paul Humphrey Award at the Carolina Population Center (USA), the International Development Research Center (Canada), and the Environmental Protection Agency (Ghana). This research received support from the Population Research Infrastructure Program awarded to the Carolina Population Center (P2C HD050924) at the University of North Carolina at Chapel Hill by the Eunice Kennedy Shriver National Institute of Child Health and Human Development and a grant from the International Development Research Centre titled “Climate Change Adaptation Research Training Capacity for Development” (106548). This research also received support from the National Science Foundation, Partnerships in International Research and Education Program (1743741).ReferencesAdeyeye SAO, Oyewole OB. 2016. An overview of traditional fish smoking in Africa. J Culin Sci Technol 14(3):198–215, 10.1080/15428052.2015.1102785. Crossref, Google ScholarBourne RRA, Stevens GA, White RA, Smith JL, Flaxman SR, Price H, et al.2013. Causes of vision loss worldwide, 1990–2010: a systematic analysis. Lancet Glob Health 1(6):e339–e349, PMID: 25104599, 10.1016/S2214-109X(13)70113-X. Crossref, Medline, Google ScholarDas I, Jagger J, Yeatts K. 2017. Biomass cooking fuels and health outcomes for women in Malawi. Ecohealth 14(1):7–19, PMID: 27800583, 10.1007/s10393-016-1190-0. Crossref, Medline, Google ScholarFlintwood-Brace A. 2016. Biomass Smoke Exposure in Traditional Smokehouses and Respiratory Symptoms among Fish Smokers at Aboadze/Abuesi in the Western Region of Ghana [master of public health thesis]. Legon, Ghana: School of Public Health, College of Health Sciences, University of Ghana. Google ScholarHankinson SE, Willett WE, Colditz GA, Seddon JM, Rosner B, Speizer FE, et al.1992. A prospective study of cigarette smoking and risk of cataract surgery in women. JAMA 268(8):994–998, PMID: 1501325, 10.1001/jama.1992.03490080068026. Crossref, Medline, Google ScholarLee AG, Kaali S, Quinn A, Delimini R, Burkart K, Opoku-Mensah J, et al.2019. Prenatal household air pollution is associated with impaired infant lung function with sex-specific effects. Evidence from GRAPHS, a cluster randomized cookstove intervention trial. Am J Respir Crit Care Med 199(6):738–746, PMID: 30256656, 10.1164/rccm.201804-0694OC. Crossref, Medline, Google ScholarSeddon JM, Willett WC, Speizer FE, Hankinson SE. 1996. A prospective study of cigarette smoking and age-related macular degeneration in women. JAMA 276(14):1141–1146, PMID: 8827966, 10.1001/jama.1996.03540140029022. Crossref, Medline, Google ScholarStone RA, Wilson LB, Ying GS, Liu C, Criss JS, Orlow J, et al.2006. Associations between childhood refraction and parental smoking. Invest Ophthalmol Vis Sci 47(10):4277–4287, PMID: 17003416, 10.1167/iovs.05-1625. Crossref, Medline, Google ScholarVan Vliet EDS, Kinney PL, Owusu-Agyei S, Schluger NW, Ae-Ngibise KA, Whyatt RM, et al.2019. Current respiratory symptoms and risk factors in pregnant women cooking with biomass fuels in rural Ghana. Environ Int 124:533–540, PMID: 30685455, 10.1016/j.envint.2019.01.046. Crossref, Medline, Google ScholarThe authors declare they have no actual or potential competing financial interests.FiguresReferencesRelatedDetails Vol. 130, No. 1 January 2022Metrics Downloaded 1,186 times About Article Metrics Publication History Manuscript received22 June 2021Manuscript revised13 November 2021Manuscript accepted7 December 2021Originally published12 January 2022 Financial disclosuresPDF download License information EHP is an open-access journal published with support from the National Institute of Environmental Health Sciences, National Institutes of Health. All content is public domain unless otherwise noted. Note to readers with disabilities EHP strives to ensure that all journal content is accessible to all readers. However, some figures and Supplemental Material published in EHP articles may not conform to 508 standards due to the complexity of the information being presented. If you need assistance accessing journal content, please contact [email protected]. Our staff will work with you to assess and meet your accessibility needs within 3 working days.}, number={1}, journal={ENVIRONMENTAL HEALTH PERSPECTIVES}, author={Weyant, Cheryl L. and Amoah, Antwi-Boasiako and Bittner, Ashley and Pedit, Joe and Codjoe, Samuel Nii Ardey and Jagger, Pamela}, year={2022}, month={Jan} }
 @article{bittner_cross_hagan_malings_lipsky_grieshop_2022, title={Performance characterization of low-cost air quality sensors for off-grid deployment in rural Malawi}, volume={15}, ISSN={["1867-8548"]}, url={https://doi.org/10.5194/amt-15-3353-2022}, DOI={10.5194/amt-15-3353-2022}, abstractNote={Abstract. Low-cost gas and particulate matter sensor packages offer a compact, lightweight, and easily transportable solution to address global gaps in air quality (AQ) observations. However, regions that would benefit most from widespread deployment of low-cost AQ monitors often lack the reference-grade equipment required to reliably calibrate and validate them. In this study, we explore approaches to calibrating and validating three integrated sensor packages before a 1-year deployment to rural Malawi using colocation data collected at a regulatory site in North Carolina, USA. We compare the performance of five computational modeling approaches to calibrate the electrochemical gas sensors: k-nearest neighbors (kNN) hybrid, random forest (RF) hybrid, high-dimensional model representation (HDMR), multilinear regression (MLR), and quadratic regression (QR). For the CO, Ox, NO, and NO2 sensors, we found that kNN hybrid models returned the highest coefficients of determination and lowest error metrics when validated. Hybrid models were also the most transferable approach when applied to deployment data collected in Malawi. We compared kNN hybrid calibrated CO observations from two regions in Malawi to remote sensing data and found qualitative agreement in spatial and annual trends. However, ARISense monthly mean surface observations were 2 to 4 times higher than the remote sensing data, partly due to proximity to residential biomass combustion activity not resolved by satellite imaging. We also compared the performance of the integrated Alphasense OPC-N2 optical particle counter to a filter-corrected nephelometer using colocation data collected at one of our deployment sites in Malawi. We found the performance of the OPC-N2 varied widely with environmental conditions, with the worst performance associated with high relative humidity (RH >70 %) conditions and influence from emissions from nearby residential biomass combustion. We did not find obvious evidence of systematic sensor performance decay after the 1-year deployment to Malawi. Data recovery (30 %–80 %) varied by sensor and season and was limited by insufficient power and access to resources at the remote deployment sites. Future low-cost sensor deployments to rural, low-income settings would benefit from adaptable power systems, standardized sensor calibration methodologies, and increased regional regulatory-grade monitoring infrastructure.}, number={11}, journal={ATMOSPHERIC MEASUREMENT TECHNIQUES}, author={Bittner, Ashley S. and Cross, Eben S. and Hagan, David H. and Malings, Carl and Lipsky, Eric and Grieshop, Andrew P.}, year={2022}, month={Jun}, pages={3353–3376} }
 @article{polimera_kannappan_richardson_bittner_ferguson_moffett_eckert_bellovary_norris_2022, title={RESOLVE and ECO: Finding Low-metallicity z similar to 0 Dwarf AGN Candidates Using Optimized Emission-line Diagnostics}, volume={931}, ISSN={["1538-4357"]}, DOI={10.3847/1538-4357/ac6595}, abstractNote={Abstract Existing star-forming vs. active galactic nucleus (AGN) classification schemes using optical emission-line diagnostics mostly fail for low-metallicity and/or highly star-forming galaxies, missing AGN in typical z ∼ 0 dwarfs. To recover AGN in dwarfs with strong emission lines (SELs), we present a classification scheme optimizing the use of existing optical diagnostics. We use Sloan Digital Sky Survey emission-line catalogs overlapping the volume- and mass-limited REsolved Spectroscopy Of a Local VolumE (RESOLVE) and Environmental COntex (ECO) surveys to determine the AGN percentage in SEL dwarfs. Our photoionization grids show that the [O iii ]/H β versus [S ii ]/H α diagram (S ii plot) and [O iii ]/H β versus [O i ]/H α diagram (O i plot) are less metallicity sensitive and more successful in identifying dwarf AGN than the popular [O iii ]/H β versus [N ii ]/H α diagnostic (N ii plot or “BPT diagram”). We identify a new category of “star-forming AGN” (SF-AGN) classified as star-forming by the N ii plot but as AGN by the S ii and/or O i plots. Including SF-AGN, we find the z ∼ 0 AGN percentage in dwarfs with SELs to be ∼3%–16%, far exceeding most previous optical estimates (∼1%). The large range in our dwarf AGN percentage reflects differences in spectral fitting methodologies between catalogs. The highly complete nature of RESOLVE and ECO allows us to normalize strong emission-line galaxy statistics to the full galaxy population, reducing the dwarf AGN percentage to ∼0.6%–3.0%. The newly identified SF-AGN are mostly gas-rich dwarfs with halo mass <10 11.5 M ⊙ , where highly efficient cosmic gas accretion is expected. Almost all SF-AGN also have low metallicities ( Z ≲ 0.4 Z ⊙ ), demonstrating the advantage of our method.}, number={1}, journal={ASTROPHYSICAL JOURNAL}, author={Polimera, Mugdha S. and Kannappan, Sheila J. and Richardson, Chris T. and Bittner, Ashley S. and Ferguson, Carlynn and Moffett, Amanda J. and Eckert, Kathleen D. and Bellovary, Jillian M. and Norris, Mark A.}, year={2022}, month={May} }
 @article{malings_westervelt_hauryliuk_presto_grieshop_bittner_beekmann_subramanian_2020, title={Application of low-cost fine particulate mass monitors to convert satellite aerosol optical depth to surface concentrations in North America and Africa}, volume={13}, ISSN={["1867-8548"]}, url={https://doi.org/10.5194/amt-13-3873-2020}, DOI={10.5194/amt-13-3873-2020}, abstractNote={Abstract. Low-cost particulate mass sensors provide opportunities to assess air quality at unprecedented spatial and temporal resolutions. Established traditional monitoring networks have limited spatial resolution and are simply absent in many major cities across sub-Saharan Africa (SSA). Satellites provide snapshots of regional air pollution but require ground-truthing. Low-cost monitors can supplement and extend data coverage from these sources worldwide, providing a better overall air quality picture. We investigate the utility of such a multi-source data integration approach using two case studies. First, in Pittsburgh, Pennsylvania, both traditional monitoring and dense low-cost sensor networks are compared with satellite aerosol optical depth (AOD) data from NASA's MODIS system, and a linear conversion factor is developed to convert AOD to surface fine particulate matter mass concentration (as PM2.5). With 10 or more ground monitors in Pittsburgh, there is a 2-fold reduction in surface PM2.5 estimation mean absolute error compared to using only a single ground monitor. Second, we assess the ability of combined regional-scale satellite retrievals and local-scale low-cost sensor measurements to improve surface PM2.5 estimation at several urban sites in SSA. In Rwanda, we find that combining local ground monitoring information with satellite data provides a 40 % improvement in surface PM2.5 estimation accuracy with respect to using low-cost ground monitoring data alone. A linear AOD-to-surface-PM2.5 conversion factor developed in Kigali, Rwanda, did not generalize well to other parts of SSA and varied seasonally for the same location, emphasizing the need for ongoing and localized ground-based monitoring, which can be facilitated by low-cost sensors. Overall, we find that combining ground-based low-cost sensor and satellite data, even without including additional meteorological or land use information, can improve and expand spatiotemporal air quality data coverage, especially in data-sparse regions.}, number={7}, journal={ATMOSPHERIC MEASUREMENT TECHNIQUES}, publisher={Copernicus GmbH}, author={Malings, Carl and Westervelt, Daniel M. and Hauryliuk, Aliaksei and Presto, Albert A. and Grieshop, Andrew and Bittner, Ashley and Beekmann, Matthias and Subramanian, R.}, year={2020}, month={Jul}, pages={3873–3892} }
 @article{palumbo_kannappan_frazer_eckert_norman_fraga_quint_amram_oliveira_bittner_et al._2020, title={Linking compact dwarf starburst galaxies in the RESOLVE survey to downsized blue nuggets}, volume={494}, ISSN={["1365-2966"]}, DOI={10.1093/mnras/staa899}, abstractNote={We identify and characterize compact dwarf starburst (CDS) galaxies in the RESOLVE survey, a volume-limited census of galaxies in the local universe, to probe whether this population contains any residual ``blue nuggets,'' a class of intensely star-forming compact galaxies first identified at high redshift $z$. Our 50 low-$z$ CDS galaxies are defined by dwarf masses (stellar mass $M_* < 10^{9.5}$ M$_{\odot}$), compact bulged-disk or spheroid-dominated morphologies (using a quantitative criterion, $\mu_\Delta > 8.6$), and specific star formation rates above the defining threshold for high-$z$ blue nuggets ($\log$ SSFR [Gyr$^{-1}] > -0.5$). Across redshifts, blue nuggets exhibit three defining properties: compactness relative to contemporaneous galaxies, abundant cold gas, and formation via compaction in mergers or colliding streams. Those with halo mass below $M_{\rm halo} \sim 10^{11.5}$ M$_{\odot}$ may in theory evade permanent quenching and cyclically refuel until the present day. Selected only for compactness and starburst activity, our CDS galaxies generally have $M_{\rm halo} \lesssim 10^{11.5}$ M$_{\odot}$ and gas-to-stellar mass ratio $\gtrsim$1. Moreover, analysis of archival DECaLS photometry and new 3D spectroscopic observations for CDS galaxies reveals a high rate of photometric and kinematic disturbances suggestive of dwarf mergers. The SSFRs, surface mass densities, and number counts of CDS galaxies are compatible with theoretical and observational expectations for redshift evolution in blue nuggets. We argue that CDS galaxies represent a maximally-starbursting subset of traditional compact dwarf classes such as blue compact dwarfs and blue E/S0s. We conclude that CDS galaxies represent a low-$z$ tail of the blue nugget phenomenon formed via a moderated compaction channel that leaves open the possibility of disk regrowth and evolution into normal disk galaxies.}, number={4}, journal={MONTHLY NOTICES OF THE ROYAL ASTRONOMICAL SOCIETY}, author={Palumbo, Michael L., III and Kannappan, Sheila J. and Frazer, Elaine M. and Eckert, Kathleen D. and Norman, Dara J. and Fraga, Luciano and Quint, Bruno C. and Amram, Philippe and Oliveira, Claudia Mendes and Bittner, Ashley S. and et al.}, year={2020}, month={Jun}, pages={4730–4750} }
 @article{richardson_polimera_kannappan_moffett_bittner_2019, title={Addressing the [O III]/H beta offset of dwarf galaxies in the RESOLVE survey}, volume={486}, ISSN={["1365-2966"]}, DOI={10.1093/mnras/stz1085}, abstractNote={Metal-poor dwarf galaxies in the local universe, such as those found in the RESOLVE galaxy survey, often produce high [O III]/Hβ ratios close to the star-forming demarcation lines of the diagnostic BPT diagram. Modelling the emission from these galaxies at lower metallicities generally underpredicts this line ratio, which is typically attributed to a deficit of photons >35 eV. We show that applying a model that includes empirical abundances scaled with metallicity strongly influences the thermal balance in HII regions and preserves the [O III]/Hβ offset even in the presence of a harder radiation field generated by interacting binaries. Additional heating mechanisms are more successful in addressing the offset. In accordance with the high sSFR typical of dwarf galaxies in the sample, we demonstrate that cosmic ray heating serves as one mechanism capable of aligning spectral synthesis predictions with observations. We also show that incorporating a range of physical conditions in our modelling can create even better agreement between model calculations and observed emission-line ratios. Together these results emphasize that both the hardness of the incident continuum and the variety of physical conditions present in nebular gas clouds must be accurately accounted for prior to drawing conclusions from emission-line diagnostic diagrams.}, number={3}, journal={MONTHLY NOTICES OF THE ROYAL ASTRONOMICAL SOCIETY}, author={Richardson, Chris T. and Polimera, Mugdha S. and Kannappan, Sheila J. and Moffett, Amanda J. and Bittner, Ashley S.}, year={2019}, month={Jul}, pages={3541–3549} }