|
|||||
|
|
Host density dependence and environmental factors affecting laurel wilt disease incidence |
|
| Authors: | Robin A. Choudhury Hong Ling Er Marc A. Hughes Jason A. Smith Gretchen E. Pruett Joshua Konkol Randy C. Ploetz James J. Marois Karen A. Garrett Ariena H. C. van Bruggen |
| Affiliation: | 1. Department of Plant Pathology, University of Florida, Gainesville, FL, USA;2. Department of Plant Pathology, University of Florida, Gainesville, FL, USA Emerging Pathogens Institute, University of Florida, Gainesville, FL, USA;3. Department of Plant Pathology, University of Florida, Gainesville, FL, USA Komohana Research and Extension Center, University of Hawaiʻi at Mānoa, Hilo, HI, USA;4. Tropical Research and Education Center, University of Florida, Homestead, FL, USA;5. North Florida Research and Education Center, University of Florida, Quincy, FL, USA;6. Department of Plant Pathology, University of Florida, Gainesville, FL, USA Emerging Pathogens Institute, University of Florida, Gainesville, FL, USA Food Systems Institute, University of Florida, Gainesville, FL, USA |
| Abstract: | Host size, density, and distribution, in addition to climate, can affect the likelihood a pathogen will invade and saturate landscapes. Laurel wilt, caused by the vector-borne pathogen Raffaelea lauricola, has devastated populations of native Lauraceae in the south-eastern USA, and continues to spread. We surveyed 87 plots in six coastal islands in South Carolina, Georgia, and north Florida, and one inland site (Archbold Biological Station) in south Florida for laurel wilt-affected and unaffected individual plants belonging to the genus Persea. The coastal island sites were surveyed once in 2008 or 2009, and the inland site was surveyed eight times from 2011 to 2013. Disease incidence per plot ranged from 0% to 96%, with mean disease incidence 42% across all sites. Disease incidence was positively correlated with trunk diameter and with density of hosts with trunk diameter >5 cm, but negatively correlated with the degree of clustering, which was highest for small trees. A recursive partitioning model indicated that higher disease incidence was associated with moderate temperatures, wider trunk diameter, lower relative humidity, and lower wind speeds. Disease progress over time at Archbold followed a Gompertz curve, plateauing at 3% in 2 years. The dispersal kernel for disease incidence from a focus followed a negative exponential distribution. The number of plots with diseased trees at Archbold was similar for redbay (P. borbonia) and swampbay (P. palustris) but was lower for silkbay (P. humilis). Understanding how host density, size, and diversity interact with environmental effects will help guide future risk prediction efforts. |
| Keywords: | classification and regression tree dispersion gradient forest disease host clustering invasive species redbay ambrosia beetle |