Accuracy of Wet Bulb Globe Temperature (WBGT) Obtained from Regional Weather Station Data for Agricultural Settings in Eastern North Carolina as Proxy for WBGT Obtained from Local Heat Stress Monitors

ECU Author/Contributor (non-ECU co-authors, if there are any, appear on document)

Danielle Nicole Dillane (Creator)

Institution

East Carolina University (ECU )

Web Site: http://www.ecu.edu/lib/

Abstract: Introduction: \nAgricultural workers are at high risk of suffering from heat stress and subsequently developing heat-related illnesses (HRIs) due to the outdoor nature of their work. This risk can be exacerbated in hot, humid temperatures such as those that define the climate found in Eastern North Carolina (ENC). It is critical for employers to monitor worker safety and assess occupational heat exposure on a daily basis so that preventative measures can be taken as needed to reduce the risk of HRIs. The Wet Bulb Globe Temperature (WBGT) Index is considered the ‘gold standard’ for assessing occupational heat stress among outdoor workers. However, traditional WBGT instrumentation may not always be readily available. Alternatively, WBGT can be calculated through models utilizing weather station data that is routinely collected and therefore readily available. Previous models have estimated WBGT based solely on standard meteorological data using heat stress data obtained from military settings. The main purpose of this study was to assess the accuracy of calculating WBGT indices from weather station data compared to instrumentation-based data obtained from locally stationed heat stress monitors in an agricultural setting. The specific aims of this study are to: 1) assess the accuracy of utilizing WBGT indices obtained from weather station data as a proxy for those obtained from instrumentation (primary aim), and 2) determine how farmers across ENC evaluate daily heat stress risk and handle/report HRI symptoms and cases among their farmworkers (secondary aim). \nMethods: \nQUESTemp°34 heat stress monitors were set up daily, weather permitting, from 8AM to 5 PM at two agricultural sites in Bertie County during the period of April-July 2021. Weather station data were collected from Weather Underground (Weather Underground, 2020) using the zip code for both study sites. The OSHA Outdoor WBGT Calculator was utilized to calculate the outdoor WBGT index for each site using the collected daily weather station data. Solar irradiance was estimated using two methods: 1) estimated solar irradiance calculated by OSHA Outdoor WBGT Calculator [weather data-based method 1/WeatherCalc1] and 2) Estimated Irradiance Based on Cloud Cover table from the heat stress chapter (Chapter 4) of OSHA’s Technical Manual (Table 1) (OSHA, 2017) [weather data-based method 2/WeatherCalc2]. The hourly average of the measured (instrumentation-obtained) WBGT index was determined and compared to the hourly calculated WBGT index using weather station data. Additionally, a survey was developed and distributed via Qualtrics to farmers across 27 counties in ENC to assess how HRI symptoms are addressed through training and how HRIs are handled and reported once they arise. The survey contained original questions developed to address the research questions of this study. At least one cooperative extension agent from each of the 41 counties in ENC were asked to provide farmer contact information as a method of recruiting participants for the survey. The study was approved by East Carolina University’s (ECU) University and Medical Center Institutional Review Board (UMC-IRB#21-000428). Line graphs and scatter plots were created to visualize the data. Bland-Altman plots were also created to assess whether the approximation behaved the same at high versus low temperatures throughout the study period. Pearson correlations were calculated to assess the strength of association between the weather station data-based and instrumentation-based WBGT indices. Statistical analysis software R (R Core Team, 2021) was used to create Bland-Altman plots and correlations. \nResults:\nLine graphs and scatter plots showed the weather data-based WBGT indices (WeatherCalc1 and WeatherCalc2) tended to be higher than instrumentation-based WBGT indices (MonitorSite1 and MonitorSite2), and that the method for calculating solar irradiance impacted WBGT indices as WeatherCalc2 WBGT indices tended to be higher than WeatherCalc1 WBGT indices. \nAll correlations were greater than 0.91. Despite a strong correlation, Bland-Altman plots suggest that utilizing zip code level weather data to calculate WBGT cannot be done accurately within ±1°C at least 90% of the time as a large portion of the data points fall outside of the limits of agreement set at ±1°C across all temperature ranges. This criterion was developed from the primary research article used for this study (Liljegren et al, 2008) which concluded accurate findings of WBGT predictions within ±1°C of the measured value 91-100% of the time, depending on location and functionality of instrumentation utilized. An overall response rate of 8.2% (147/1,788) was recorded for the survey. Survey results indicated that more widespread heat stress training is needed among agricultural workers, rest breaks with provided water need to be mandatory, a system needs to be in place for reporting and handling HRIs on individual farms, and more farmers need to assess risk of HRIs daily by checking either the heat index or WBGT index.\nConclusion: \nOverall, utilizing zip code level weather data to accurately calculate WBGT indices that are within ±1°C of local instrument-measured WBGT indices cannot be accomplished at least 90% of the time, for either method used to calculate solar irradiance, using the model developed by Liljegren et al. (2008), thus answering the first primary research question for this study. For monitoring site 1, methods 1 and 2 of calculating weather data-based WBGT indices were accurate within ±1°C of instrumentation-based WBGT indices 45.0% and 30.5% of the time (respectively) for the entire study period. For monitoring site 2, methods 1 and 2 of calculating weather data-based WBGT indices were accurate within ±1°C of instrumentation-based WBGT indices 33.7% and 23.6% of the time (respectively) for the entire study period. This combined information answers the second primary research question for this study. However, the weather data-based WBGT indices were higher than instrumentation-based WBGT indices, indicating that their use may result in an overestimation of the risk of heat stress to workers. From a public health standpoint, the conservative and protective nature of overestimating WBGT indices utilizing weather data could be lifesaving. It is strongly recommended that all farmers across ENC provide annual heat stress training to their employees, implement a system for reporting and handling HRIs on their farm, provide water and mandatory rest breaks to employees, and check the heat index or WBGT index daily to assess heat exposure risks. These changes will directly provide a safer work environment for outdoor agricultural workers and could ultimately impact public health by lowering incidences of HRIs across ENC.

Additional Information

Publication

Dissertation

Language: English

Date: 2023

Subjects

heat stress;WBGT;outdoor agricultural workers;OSHA WBGT outdoor calculator;worker safety;occupational exposure;exposure assessment

Email this document to