Leaf and Soil Nutrient Relations of Elaeocarpus joga Merr. in Oceanic Island Calcareous Soils

The leaf nutrient status and stoichiometry of nitrogen (N), phosphorus (P), and potassium (K) were determined for Elaeocarpus joga trees in Guam’s dominant calcareous soils to understand nutrient limitations in limestone soils of Oceania and contribute to global databases on leaf economic spectrum studies. Supplemental N, P, or K was added to soils to determine plant growth and nutrient concentration responses. Leaf and soil quantifications of nutrients enabled multiple trait comparisons. Supplemental N stimulated growth of young cultivated plants without affecting leaf N concentrations. Supplemental K increased leaf K concentration but did not generate a growth response. Supplemental P did not affect growth or leaf P concentration. N:P, N:K, andK:Pweremost influenced byK additions. Leaf N and P concentrations of mature trees in unmanaged settings were similar to unfertilized young plants in the controlled study, but leaf K concentration was greater in the mature trees. Leaf nutrient relations were not overtly related to soil nutrient relations for mature trees. Results indicate that N and K are the limiting factors in calcareous soils of the Mariana Islands for this endemic tree species, age and size of trees do not greatly influence leaf nutrient content, and leaf stoichiometry is constrained and less variable than soil stoichiometry. Studying the N, P, and K relations in terrestrial plant leaves can provide useful information about nutrient limitations in dominant soils and the interactions among species (Ågren, 2008; Sterner and Elser, 2002). Stoichiometry may be useful for interpretation, and is defined as the relationship between two constituents of the ionome or metabolome and expressed as a quotient. For example, numerous authors have employed robust data sets to suggest that quantifying variations in leaf N, P, and K may be highly informative. These studies collectively indicate that leaf N:P above 16 suggests P limitation to plant biomass production, N:P below 14 suggests N limitation, and N:K above 2.1 or K:P below 3.4 indicate K limitation (Aerts and Chapin, 2000; G€usewell, 2004; Koerselman and Meuleman, 1996; OldeVenterink et al., 2003;Tessier and Raynal, 2003). Other reports illuminate the importance of K concentrations in developing an understanding of tree physiology, and how N:K and K:P correlate with climate gradients (Gotelli et al., 2008; Rivas-Ubach et al., 2012; Sardans et al., 2012). Continued research into these generalities may help to answer critical questions about nutrient limitations at the local scale, and enhance our understanding of relationships between stoichiometric shifts and environmental gradients on the global scale. All natural and managed systems may not conform to the general expectations of nutrient stoichiometry in plant leaf tissue (Craine et al., 2008). Therefore, caution should be used before imposing sweeping interpretations of widespread plant leaf nutrient relations until those relationships have been sufficiently studied at the local level. Moreover, underrepresented geographic locations may provide useful local results to improve the global understanding of terrestrial plant nutrient limitations and function, which may improve interpretations of global integrations such as leaf economic spectrum dynamics (Donovan et al., 2011; Metali et al., 2012; Wright et al., 2004, 2005). Recent developments have exposed the potential of how anthropogenic changes may be altering elemental cycling. The continuous increase in global populations and highly variable growth conditions brought about by climate change underscore a critical need to better understand the management of plant productivity (vonWir en, 2011). These global change factors also highlight the urgency Fig. 1. Elaeocarpus joga trees possess highly distinctive traits. (A) The tree canopy is known for its unique architecture and foliage texture. (B) The distinctive flower corolla generates the nickname fairy petticoats. (C) Mature fruits are highly apparent in the canopy or on the forest floor after abscission because the exocarp transmits iridescent blue by light refraction rather than by pigmentation. (D) Cleaned endocarps are worn or held as beads by Hindu and Buddhist followers to provide myriad spiritual benefits. Received for publication 16 June 2015. Accepted for publication 17 Sept. 2015. This material is based on work that is supported by U.S. Department of Defense Agreement DAMD17-00-2-0021 and by the National Institute of Food and Agriculture, the U.S. Department of Agriculture McIntire Stennis project under 2013-32100-06057. Corresponding author. E-mail: tmarler@uguam. uog.edu. 1644 HORTSCIENCE VOL. 50(11) NOVEMBER 2015 inadequately understanding and documenting the traits of highly managed systems, minimally managed agroforest systems, and natural ecosystems. Guam is the southernmost island within the Mariana Island Archipelago and largest island within Micronesia. Two critical anthropogenic threats have emerged during the last quarter century to threaten native tree species. First, the proliferation of Boiga irregularis following its invasion during the late 1940s by way of shipment of the U.S. military equipment (Rodda et al., 1992) resulted in extirpation of bird and bat species that has led to devastating impacts on the natural forest and agricultural ecosystems (Fritts and Rodda, 1998; Rogers et al., 2012). Second, the continued loss of vegetated habitats due to land conversion magnified pressures on the remaining fragmented habitats throughout the island. The amplified impact frommultiple new alien arthropod species and an expansive military buildup will bring even greater need for understanding how tomanage the urban forest (Marler, 2013a, 2013b;Marler and Moore, 2011). Elaeocarpus joga is a distinctive member of the limestone forests of the Mariana Islands, and provides a striking canopy in the urban forest (Fig. 1). This and other members of the genus produce iridescent blue fruits, with ultrastructural light refraction generating the blue phenotype (Lee, 1991). Cleaned endocarps provide the resource for Rudraksha, which carries myriad medicinal and spiritual symbolisms in Hinduism and Buddhism (Seetha, 2008). This unique tree is one of the forest species that have become rare in recent decades due to habitat loss, increasing fragmentation, loss of avian seed dispersers, and impacts of alien ungulates (Raulerson and Rinehart, 1991). Leaf chemical traits of arborescent native forest species have not been studied for this portion of Oceania, so an understanding of nutrient cycling and stoichiometry for the region is lacking. A better understanding of the nutrient dynamics of tree species will benefit managers working to develop longrange management plans for this and other native forest trees (Wiles, 2005). We employed E. joga as a case study to begin developing an understanding of nutrient limitation and stoichiometry in the Mariana Islands. We used supplemental N, P, or K in a controlled system and the relationship between leaf and soil nutrient levels of unmanaged mature trees to meet this objective. Materials and Methods Fertilization study. Container-grown E. joga seedlings were employed to initiate a field study in Yigo, Guam. This field was characterized with Guam soil series, consisting of shallow cobbly clay loams that formed in sediments that overlie coralline limestone (Clayey, gibbsitic, nonacid, isohyperthermic, and Lithic Ustorthents; Young, 1988). This is the primary substrate forGuam’s contemporary E. joga populations. Soil at the site exhibited the following characteristics: organic matter = 19.3%, pH = 7.7, total N = 9.8mg·g, MehlichIII P = 163 mg·g, Mehlich-III K = 222 mg·g, Mehlich-III magnesium (Mg) = 477 mg·g, and Mehlich-III calcium (Ca) = 7080 mg·g. The plants were established in 2.6-L containers growing in a 60% peat and 40% perlite medium. Nursery fertilization consisted of weekly application of 50 mL of complete fertilizer solution (Excel; GraceSierra, Milpitas, CA) providing 7.5 mM N. In addition, the plants received a monthly application of 50 mL of iron (Fe) chelate (Sprint 138; Becker Underwood, Ames, IA). Our use of low-dosage frequent-soluble fertilizer applications in the nursery produced plants with minimal nutrient reserves in the container media after the plants were transplanted. Fig. 2. Growth traits of Elaeocarpus joga plants grown in Guam cobbly clay soils as influenced by fertilization with nitrogen (N), phosphorus (P), or potassium (K) to achieve 50or 100 g per plant per year. (A) Increase in basal stem diameter, (B) increase in plant height, and (C) increase in canopy diameter. Mean ± SE, n = 10. HORTSCIENCE VOL. 50(11) NOVEMBER 2015 1645 The plants were transplanted into a 2 · 4m grid in the field on 13 Feb. 2001 when they were 85 ± 3 cm tall with canopy diameter of 37 ± 3 cm and basal stem diameter of 11 ± 1 mm. We used a randomized complete block design positioned on a north-south gradient. Ten single-tree replications were assigned to each prescribed treatment. One treatment consisted of control trees, which received no supplemental fertilizer. Fertilization of N was performed with NH4NO3, fertilization of P was performed with Ca(H2PO4)2, and fertilization of K was performed with K2SO4. There were two dose levels per fertilizer treatment. One dose provided 50 g of the element per tree per year, and one dose provided 100 g of the element per tree per year. For N and K, the annual rate was split among fertilizer applications made on the day of planting and every 2 month thereafter. For P, the entire amount of fertilizer was incorporated into the planting hole and backfill on the date of planting. Each plant received irrigation every 3 d by way of drip emitters that delivered 7.5 L per plant such that each plant received 17.5 L·week. Mature green leaf samples were collected from the control trees and the trees receiving the 100 g dose level for each of the nutrient addition treatments (trees in the 50 g dose level were not included). Samples were restricted to recently expanded and sunexposed leaves, and each tree’s composite sample consisted of at least 15 leaves. These leaf samples were collected on 28 Nov. 2001, which enabled collection at the end of Guam’s annual rainy season and a delay of 4 weeks after the most recent fertilizer applications. The samples were dried to constant weight at 70 C, milled, then total N was determined by dry combustion (FLASH EA1112 CHN analyzer; Thermo Fisher, Waltham, MA), and P and K were determined by inductively coupled plasma optical emission spectrometry (Spectro Genesis; SPECTRO Analytical Instruments, Kleve, Germany) (Hou and Jones, 2000). Final tree growth measurements were conducted on 4 Feb. 2002 following 1 year of growth under the experimental conditions. Terminal measurements included basal stem circumference measured with a flexible tape measure then converted to diameter, and plant height and diameter measured with a rigid tape measure. Soils and leaves from natural habitats. Ten mature E. joga trees were visited from 3 to 11 Nov. 2014. The trees ranged in height from 8 to 14 m, with a mean of 10 m. They ranged in diameter at breast height from 41 to 53 cm, with a mean of 48 cm. For each tree, youngest fully expanded green leaves were harvested from the cardinal positions within the canopy using a telescopic pole pruner. Soil samples were composed of a composite from the surface to the solid CaCO3 contiguous base, and ranged in depth to 11–18 cm of soil. Samples were obtained from eight locations at the edge of the canopy. All leaves from each tree were combined into one composite tissue sample. All soil samples at each tree site were combined to create one composite soil sample. The leaves were dried at 70 Candmilled, and analyzed as previously described with the addition of carbon (C) determined by dry combustion. The soil was dried at 50 C then C and N were determined by dry combustion, available Pwas determined as Olsen-P (Olsen et al., 1954), and K, Mg, Ca, Fe, manganese (Mn), copper, and zinc were determined as extractable content by diethylene triamine pentaacetic acid digestion. Data handling. The growth defined by tree height, canopy diameter, and basal stem diameter during the 1-year study was calculated by subtracting the initial measurement from the terminal measurement for each response variable. The N:P, N:K, and K:P data were log transformed to satisfy preconditions of analysis of variance (ANOVA). Other data required no transformations. Growth and leaf nutrient variables were subjected to ANOVA within the randomized complete block design using SAS V9.2. The relationship between various pairs of response variables were determined by regression analysis using the PROC GLM statement of SAS V9.2.