Scientific Notes 1247 EXPANSION AND IMPACT OF LAUREL WILT IN THE FLORIDA EVERGLADES 1, 2 3 LEROY RODGERS *, ANDREW DERKSEN AND TONY PERNAS 1South Florida Water Management District. Land Resources Bureau, West Palm Beach, FL 33406, USA 2Florida Department of Agriculture and Consumer Services, Division of Plant Industry, Miami, FL 33158, USA 3National Park Service, Florida/Caribbean Exotic Plant Management Team, Palmetto Bay, FL 33157, USA *Corresponding author; E-mail: [email protected] Tree islands are a prominent feature of the Aerial surveys were conducted during Mar– Florida Everglades landscape. These ovoid patch- May 2011, and May–Jun 2013 to determine the es of forest vegetation are scattered throughout distribution and abundance of the wilt in and ad- a matrix of herbaceous marsh and are important jacent to the Everglades. Tree islands displaying contributors to Everglades biodiversity and fun- symptomatic trees were recorded using a digital damental ecosystem processes. Tree islands have aerial sketch mapping system during planned high species diversity, provide wading bird nest- invasive plant surveys (see Rodgers et al. 2014). ing habitat, and provide critical upland habitat Observers recorded presence and abundance of for wildlife during the wet season. Everglades symptomatic swamp bay trees along fixed flight tree islands also play a key role in biogeochemical transects spaced 1-km apart, allowing for com- cycling and serve as nutrient sinks, particularly plete coverage of the survey area (Fig. 1). Ob- in the case of phosphorus (Wetzel et al. 2005). As servers digitized clusters of symptomatic trees as such, tree island protection and restoration are polygons using field mapping software (GeoLink® increasingly viewed as priorities in the national SketchmapperTM) and estimated percent cover of Everglades restoration effort (Sklar & Van der canopy damage for each polygon. Data were im- Valk 2003). ported into a GIS (ESRI® ArcGIS Desktop™) to Swamp bay (Persea palustris (Raf.) Sarg.) measure the proportion of tree island canopy with (Laurales: Lauraceae) is a common tree species laurel wilt symptoms and range of occupancy of Everglades tree islands (Heisler et al. 2002), (generalized area containing symptomatic trees but ranges from dominant to absent in individu- within the survey area). Digital aerial imagery al islands (Wetzel et al. 2008; Engel et al. 2009; and tree island species composition data (Engel et Brandt et al. 2003). The tree also carries cultural al. 2009) were used to estimate the proportion of significance in the Everglades as its bark and tree islands with swamp bay affected by the wilt. leaves are key reagents in traditional Seminole Within 2 months of both surveys, sapwood medicine (Snow & Stans 2001). Swamp bay is one samples were collected from a subset of symptom- of several members of the Lauraceae affected by atic trees and submitted for testing to the FDACS laurel wilt in the southeastern USA. Laurel wilt Plant Pathology Laboratory in Gainesville, Flor- is a vascular wilt disease caused by Raffaelea lau- ida. Fungi were isolated from samples following ricola T.C. Harr. (Ophiostomatales: Ophiostoma- Fraedrich et al. (2008), and confirmed using the taceae), a symbiont of its vector, the exotic redbay morphological and molecular methods described ambrosia beetle, Xyleborus glabratus Eichhoff in Harmon & Brown (2009). (Coleoptera: Curculionidae: Scolytinae) (Fraed- During the 2011 aerial survey, symptomatic rich et al. 2008). trees were observed within a 4,925 ha range of Xyleborus glabratus was first collected in the occupancy in the eastern Everglades (Fig. 1). The United States near Port Wentworth, Georgia in greatest concentration of symptomatic trees oc- 2002 (Rabaglia et al. 2006), and by 2004, laurel curred in tree islands and artificially elevated wilt was associated with extensive mortality of sites (ditch spoil mounds) within partially drained redbay (P. borbonia (L.) Spreng.) (Fraedrich et wetlands adjacent to ENP. Two tree islands with- al. 2008). Since then, the vector and its phyto- in ENP harbored symptomatic trees. Sampling of pathogenic symbiont have spread throughout the 46 of the symptomatic trees confirmed that 80.5% southern Atlantic coastal plain (USDA 2014), with tested positive for R. lauricola. The 2013 survey localized losses of redbay populations exceeding revealed a substantial increase in the range of 80% (Fraedrich et al. 2008). The beetle was first occupancy (133,740 ha) (Fig. 1). Laurel wilt had detected in Miami-Dade County just outside Ever- spread further west into ENP, Water Conserva- glades National Park (ENP) in Feb 2010 (Griffiths tion Area (WCA) 3, and Big Cypress National & Derksen 2010), and laurel wilt was confirmed on Preserve. The westerly expansion is not unex- swamp bay only a few miles away from the initial pected given the prevailing winds from the east capture in February 2011 (Ploetz et al. 2011). and southeast, especially in the spring and fall 1248 Florida Entomologist 97(3) September 2014 Fig. 1. The distribution and proportions of the canopy occupied by trees with symptoms of laurel wilt in Ever- glades natural areas (2011 and 2013). when adult beetle emergence peaks (Brar et al. neighbor analysis of tree islands with symptom- 2012; Maner et al. 2013). Symptomatic swamp atic trees revealed that the mean minimum dis- bay were observed in 322 tree islands and along tance between symptomatic tree islands was 1.0 ditches, canals, and roads within the survey area. km and the maximum nearest neighbor distance Raffaelea lauricola was recovered from 93.3% of was 11.8 km. the samples taken from symptomatic trees at 15 In 2013, tree islands with at least 1 symptom- locations. atic swamp bay had a median canopy damage of The western edge of the laurel wilt focus ex- 5.0% (interquartile range 2.5 to 10.0%). An analy- panded by an average of 53.3 ± 1.5 SE km be- sis of tree island canopy composition (Engel et al. tween 2011 and 2013. Assuming a constant rate 2009) within the 2013 range of occupancy indi- of expansion during the 2 yr, laurel wilt spread cates that swamp bay importance values (sum of across the east-central Everglades at 26.6 km/ relative tree density and basal area) range from yr, which falls in the lower range of the Koch & 0 to 43.4 (mean = 3.8), suggesting a wide range Smith (2008) estimates (24 to 55 km/yr). Koch of potential canopy loss among tree islands. Tree & Smith’s expansion estimates were associated islands with high swamp bay importance values with highly variable habitats in more temper- are expected to sustain substantial canopy loss, ate environments and involved a different host as observed for some tree islands in this survey species, redbay. Although the present data were (maximum estimated canopy damage = 50%). from a relatively uniform subtropical habitat and Within the current range of occupancy, 76% another host species, swamp bay, they may pro- of the tree islands known to include swamp bay vide a better estimate of how quickly laurel wilt were not yet symptomatic, suggesting that lau- can spread in the absence of anthropogenic move- rel wilt will cause additional mortality. We note ment. The clustered distribution of tree islands that the disease was rare in the southwest por- across the Everglades landscape may also provide tion of Water Conservation Area 3 (WCA 3) (Fig. insight into the capacity of R. lauricola vectors 1) where high importance values of swamp bay to naturally disperse long distances. A nearest would predict the presence of laurel wilt. Factors Scientific Notes 1249 associated with this apparent low rate of wilt de- por los P. palustris con sintomos de la marchitez velopment should be identified. del laurel en las islas de árboles en los Evergla- The response of Everglades tree islands to sud- des en marzo del 2011 y junio del 2013 utilizando den canopy loss is difficult to predict and warrants un sistema para hacer mapas de imágenes aéreas future evaluation. An abrupt loss of tree island digitales. El rango geográfico ocupado por la en- canopy where swamp bay is an important com- fermedad incrementó de 4,925 a 133,740 hectá- ponent could reduce peat accretion rates (Cahoon reas durante este período de 26 meses. El daño et al. 2003) and lead to reduced tree island eleva- del dosel fue altamente variable de una pérdida tion and stability (Wetzel et al. 2005). Canopy del 0 a 50 % del dosel. Sin embargo, los datos de disturbance also influences the ease with which la composición de especies de árboles en las islas communities are invaded (Eschtruth & Battles sugieren que probablemente hay mas mortalidad 2009). Lynch et al. (2009) reported increased colo- de P. palustris en los Everglades. La pérdida rá- nization of Everglades tree islands by the exotic pida de P. palustris en las islas de árboles podría Old World climbing fern (Lygodium microphyl- resultar en la inestabilidad a nivel de ecosistemas lum (Cav.) R. Br.; Polypodiales: Lygodiaceae) af- y la invasión de especies exóticas de plantas. ter hurricane-induced canopy disturbances and Palabras Claves: isla árbol, hacer mapas de Brazilian pepper (Schinus terebinthifolius Raddi; imágenes aéreas digitales, Persea palustris, Ra- Sapindales: Anacardiaceae) colonizes canopy ffaelea lauricola, Xyleborus glabratus gaps and eventually dominates heavily impacted tree islands (Rodgers et al. 2014). Everglades REFERENCES CITED land managers are encouraged to increase inva- sive plant monitoring and control efforts in tree BRANDT, L. A., ECKER, D., GOMEZ RIVERA, I., TRAUT, A., islands where substantial canopy has been lost AND MAZZOTTI, F. J. 2003. Wildlife and vegetation of due to laurel wilt. bayhead islands in Arthur R. Marshall Loxahatchee National Wildlife Refuge. Southeastern Nat. 2:179- 194. SUMMARY BRAR, G. S., CAPINERA, J. L., MCLEAN, S., KENDRA, P. E., PLOETZ, R. C., AND PEÑA, J. E. 2012. Effect of Laurel wilt, a vascular wilt disease caused by trap size, trap height and age of lure on sampling the vascular mycopathogen Raffaelea lauricola Xyleborus glabratus (Coleoptera: Curculionidae: T.C.