An Examination of Seasonal Influences of Environmental and Social-Economic Conditions on COVID-19 Pandemic in Coastal Virginia, U.S.A.: A Geospatial and Machine Learning Approach

Keywords: Seasonal Influences, Environmental and Social-Economic Conditions, COVID-19, Public Health
Abstract Type: Paper Abstract

Abstract

The coronavirus (COVID-19) has posted disastrous impact in most countries of the world in the past few years. Plenty of studies have been documented to examine any possible factors contributing to the outbreaks, to predict the trends of the pandemic, and to evaluate its impacts on the society. However, there are still limited records to assess the seasonal influences of physical and social conditions on COVID-19 pandemic at a refined scale, as part of a larger trend of innovation in health recovery for any state. This study aims to measure the seasonal influences of environmental and social-economic conditions on COVID-19 Pandemic in coastal Virginia, U.S.A. on a pixel basis (30×30 m). An innovative approach to public health studies was used by combining state-of-the-art geospatial technologies and machine learning. The environmental factors used for evaluation include Normalized Difference Vegetation Index, Normalized Difference Built-up Index, average temperate, and total precipitation at seasonal level. Social-economic factors include household income, poverty level, education background, and road connectivity. The possible influence of each variable on the spread of COVID-19 was examined by machine learning approaches. The most influential variables are used to assess the risk areas of COVID-19 in coastal Virginia. Data visualization offers an effective means of communicating data to scientists, policy makers, and the public. Accordingly, governments can build more resilient and healthier communities by immediately establishing communication hubs, efficiently allocating resources, and effectively easing disease burdens.