Citizen science contributes to our knowledge of invasive plant species distributions
Product Type:Journal Article
Author(s):Crall, Alycia W.; Jarnevich, Catherine S.; Young, Nicholas E.; Panke, Brendon; Renz, Mark; Stohlgren, Thomas
Suggested Citation:Crall, A. W., C. S. Jarnevich, N. E. Young, B. J. Panke, M. Renz, and T. J. Stohlgren. 2015. Citizen science contributes to our knowledge of invasive plant species distributions. Biological Invasions 17:2415-2427.
Citizen science is commonly cited as an effective approach to expand the scale of invasive species data collection and monitoring. However, researchers often hesitate to use these data due to concerns over data quality. In light of recent research on the quality of data collected by volunteers, we aimed to demonstrate the extent to which citizen science data can increase sampling coverage, fill gaps in species distributions, and improve habitat suitability models compared to professionally generated data sets used in isolation. We combined data sets from professionals and volunteers for five invasive plant species (Alliaria petiolata, Berberis thunbergii, Cirsium palustre, Pastinaca sativa, Polygonum cuspidatum) in portions of Wisconsin. Volunteers sampled counties not sampled by professionals for three of the five species. Volunteers also added presence locations within counties not included in professional data sets, especially in southern portions of the state where professional monitoring activities had been minimal. Volunteers made a significant contribution to the known distribution, environmental gradients sampled, and the habitat suitability of P. cuspidatum. Models generated with professional data sets for the other four species performed reasonably well according to AUC values (>0.76). The addition of volunteer data did not greatly change model performance (AUC > 0.79) but did change the suitability surface generated by the models, making them more realistic. Our findings underscore the need to merge data from multiple sources to improve knowledge of current species distributions, and to predict their movement under present and future environmental conditions. The efficiency and success of these approaches require that monitoring efforts involve multiple stakeholders in continuous collaboration via established monitoring networks.