COS 96-5 - Allergenic pollen in cities: Using remote sensing to determine source plant locations

Wednesday, August 9, 2017: 9:20 AM
E146, Oregon Convention Center
Daniel W. Katz and Stuart Batterman, School of Public Health, University of Michigan, Ann Arbor, MI

Pollen allergies afflict over 35 million Americans, leading to both allergic rhinitis (hay fever) and asthma attacks. Despite the considerable public health consequences of allergenic pollen, little is known about how pollen concentrations vary within cities. This is in part because there are few comprehensive maps of plants in urban areas, especially for annual weeds. If such information were available it could serve as the basis for pollen dispersion models, which could in turn produce individual level estimates of allergenic pollen exposures.

In this study, we investigated the presence and abundance of common ragweed (Ambrosia artemisiifolia) plants in Detroit, MI. To do so, we conducted a field survey in which we recorded the coordinates and potential pollen production of individual ragweed plants. We then quantified the occurrence of several land cover types that we expected to be associated with ragweed (e.g., disturbed soil). This was done by applying a supervised classification algorithm to a time series of high resolution aerial photos.


Over the course of our field survey, we took measurements on 2203 individual plants, and mapped the approximate location of 26,468 plants in Detroit. Our analysis shows that ragweed plants were associated with areas that had been classified as bare dirt within the previous several years; this was often due to building demolition. Areas which had been disturbed longer ago were also less likely to contain ragweed. This is explained by ragweed’s status as an early successional species, and highlights the need for different post-demolition land management approaches.

High resolution aerial photography and satellite imagery is becoming increasingly available in urban areas. This approach of combining remote sensing data and field surveys therefore has the potential to be applied more widely for plants of health or economic importance. In this case, high accuracy plant location data will allow for city-wide estimates of ragweed pollen concentrations, which can be linked with respiratory health outcomes