ARTICLE Nutrient Limitation in Two Everglades Tree Species Planted on Constructed Tree Islands Suresh C. Subedi & Michael S. Ross & Leonard J. Scinto Received: 22 March 2012 / Accepted: 9 October 2012 / Published online: 18 October 2012 # Society of Wetland Scientists 2012 Abstract The Everglades is a low-nutrient ecosystem occupied by marsh plant species adapted to low avail- ability of phosphorus. Recently, however, tree islands that are scattered throughout the marsh have been recognized as biogeochemical hotspots. The goal of this study was to determine the general patterns of response by common tree species when conditions limiting to optimal growth were improved by fertilization in an experimentally con- structed and managed Everglades wetland. Thirty-six trees of two species, Annona glabra and Chrysobalanus icaco, were randomly selected on two peat- and two limestone-based islands. Each tree was treated with one of three nutrient regimes: Nitrogen (N), Phosphorus (P), or Control (no addition of nutrients). Positive highly significant P-treatment effects on leaf total P and leaf N:P were observed in both species in comparison to Control trees, but neither species exhibited a similar response to N-fertilization. However, among the two species, only A. glabra responded to P-fertilization with increased growth. Both fertilized and unfertilized trees of each species exhibited a highly significant growth re- sponse to hydrological condition, with growth enhanced on less persistently flooded sites. Our experimental results identify a clear difference in species growth responses to substrate type in the two species, but do not support the idea that a single critical N:P ratio can be used to indicate nutrient limitation for all wetland trees. Keywords Everglades tree islands . Fertilization experiment . Nutrient limitation . Leaf N:P . δ 13 C . δ 15 N Introduction The Everglades is a low-nutrient system, particularly with regard to the availability of phosphorus (Noe et al. 2001; Wetzel et al. 2005). However, within this vast wetland, tree islands have been recognized as biogeochemical hotspots (Wetzel et al. 2005; Ross et al. 2006; Ross and Sah 2010; Espinar et al. 2011). In the heads of the tree islands, under tropical hardwood forest (hammock) cover, soil total phos- phorus (P) concentration is extraordinarily high in compar- ison to the surrounding marsh soils (Wetzel et al. 2008; Ross and Sah 2010). In contrast, soil total nitrogen (N) is reported to be highest in flooded areas and decreases to- wards the higher elevations (Jayachandran et al. 2004). Several P accumulation mechanisms, including deposition of aerosols, precipitation, movement in groundwater through transpiration stream, deposition of guano by birds and animal feces, creation of middens full of phosphorus- rich bones by pre-Columbian humans, and bedrock miner- alization (Ross et al. 2006; Givnish et al. 2008; Graf et al. 2008) have been demonstrated, but their relative impor- tance is currently debated. In contrast, the total nitrogen in tree island soils primarily depends on soil organic matter content and its interaction with hydrology (Jayachandran et al. 2004). Differences in hydroperiod and nutrient accumu- lation with the elevation in the tree islands may lead to relative N-limitation at upslope positions, and P-limitation downslope. Biomass production and nutrient dynamics on the tree island may also be influenced by geomorphology, as the interaction of underlying substrate with hydrology is a fun- damental driver of variation in nutrient availability among sites. There are two distinct categories of tree islands found in the Everglades based on underlying substratum: those composed entirely of peat and those that developed on shallow soils above limestone bedrock (hereafter referred S. C. Subedi (*) : M. S. Ross : L. J. Scinto Florida International University, Miami, FL, USA e-mail: ssube001@fiu.edu Wetlands (2012) 32:1163–1173 DOI 10.1007/s13157-012-0346-0