NPK Fertilization of Serianthes Plants Influences Growth and Stoichiometry of Leaf Nutrients

Abstract

The genus Serianthes has not received adequate research attention, leaving large gaps in the knowledge required to inform conservation decisions. For example, nutrient management protocols are not understood due to lack of research. Serianthes grandiflora, Serianthes kanehirae, and Serianthes nelsonii plants were grown in container culture to determine the influence of increasing edaphic nitrogen (N), phosphorus (P), or potassium (K) content on stem growth and leaf nutrient relations. Addition of N alone increased leaf N, stimulated stem height and diameter growth, increased leaf number, and reduced leaf tissue concentrations of most nutrients including P and K. Addition of K alone increased leaf K, did not influence stem growth, did not reduce N or P concentration, but caused substantial changes in leaf tissue stoichiometry. Addition of P alone increased leaf P, did not influence stem growth, did not reduce leaf N or K concentration, and exerted minimal influence on concentrations and stoichiometry of other nutrients. The results indicate that single element P fertilization may be used to mitigate P deficiencies of Serianthes saplings without the risk of causing substantial nutrient imbalances. This knowledge may be used to inform nutrient management decisions in Serianthes conservation projects.

1. Introduction

Serianthes grandiflora (Benth.) F. Müller from the Philippines, Serianthes kanehirae Fosberg from Palau and Yap, and Serianthes nelsonii Merr. from the Mariana Islands are three closely related taxa that exhibit similar growth and development when grown under the same conditions [1,2,3,4]. Serianthes nelsonii is listed as critically endangered under the International Union for Conservation of Nature [5] and endangered under the United States Endangered Species Act [6]. The federal listing places constraints on research protocols, therefore the other two species have been used as comparable taxa or as surrogates in recent studies as a means of improving S. nelsonii conservation knowledge [2,3,7,8].

Manipulations of individual macronutrients in growing systems cause numerous interactive and contextual influences on plant nutrient budget responses [9]. Fabaceae tree species have access to nitrogen fixed by endosymbionts, and for this reason their responses to experimental fertilization treatments may be unique [10,11,12]. Although the influence of nutrient availability on growth and chemical composition of the more common Fabaceae tree species has been extensively studied, protocols for managing mineral nutrition of less common taxa such as Serianthes Benth. species have not received adequate research attention.

Single macronutrient fertilizers are used in crop production systems to eliminate targeted deficiencies or to experimentally manipulate plant nutrition to identify which nutrients are most limiting. This practice may lead to nutrient imbalances because changes in elemental tissue concentrations are not homogeneous among the nutrients [13,14]. For example, growth stimulation which correlates with the added nutrient may dilute other nutrients such that they become more limiting. Biomass concentrations of these other nutrients may be reduced even more as a result of the increase in relative carbon (C) concentrations. Imbalances may also result from direct changes in relations among the minerals.

In situ S. nelsonii trees are primarily limited by phosphorus (P) and secondarily limited by potassium (K) [15]. To my knowledge, only one study has been published regarding the use of horticultural fertilization methods for S. nelsonii plants [16]. In this study, foliar sprays of dilute nutrient solutions were employed to successfully supply leaves with essential plant nutrients. Nutritional requirements have not been studied for S. grandiflora or S. kanehirae. More fertilizer studies may improve our understanding of how to manage mineral nutrition to increase conservation successes of all Serianthes taxa. The current study used S. grandiflora, S. kanehirae, and S. nelsonii plants to investigate the impact of N, P, or K fertilization on stem growth and foliar concentrations of five macronutrients and five micronutrients.

2. Materials and Methods

2.1. Serianthes kanehirae

A study was conducted in Mangilao, Guam (13.43257 N, 144.79928 E; 68 m asl) from 10 April 2015 until 11 July 2015 with S. kanehirae plants. The container substrate was soil collected from northeast Guam from the soil series that supports the S. nelsonii population (Clayey-skeletal, gibbsitic, nonacid, isohyperthermic Lithic Ustorthents). This karst soil was mixed 50:50 with quartz sand to improve drainage. Containers were 2.6 L in volume, 13 cm in height, and exhibited a surface area of 201 cm². Control plants received no fertilizer. The plants which received N, P, or K fertilizer were supplied with the equivalent of 100 kg of nutrient per ha (e.g., [17]). The N fertilizer treatment comprised two parts NH₄NO₃ and one part CO(NH₂)₂ such that each plant received ammonium, nitrate, and urea sources of N in equal amounts. The P fertilizer treatment was Ca(H₂PO₄)₂. The K fertilizer treatment was K₂SO₄. All of the P and half of the N and K dosages were mixed into the container substrate. For N and K, one half of the dosage was top-dressed on 24 May 2015.

Serianthes kanehirae seedlings growing in tubes 5 cm diameter and 12 cm deep were bare-rooted and transplanted one per container. The beginning plant size was 26 ± 1 cm tall (mean ± SE), 4.9 ± 0.2 mm stem diameter, and with 9 ± 1 leaves. The plants were arranged in a completely randomized design on a nursery bench under 50% shadecloth. They were irrigated by hand and as needed. Exposure of the nursery bench to rainfall allowed the use of scouting to determine when irrigation was necessary and was usually two to three times per week. There were six replications.

Final measurements were made on 10 July 2015 after 13 weeks of growth. Stem height and basal diameter were directly measured with a meter stick and caliper, and leaf number was counted. Fully expanded leaves were removed from the apical 10 cm of growth for tissue analysis and were dried for 48 h at 75 °C in a forced draft oven then milled to pass through 20-mesh screen. Total C and N concentration was determined by dry combustion (FLASH EA1112 CHN Analyzer, Thermo Fisher, Waltham, MA, USA). The samples were also digested by a microwave system with nitric acid and peroxide, then other elements were quantified by inductively coupled plasma optical emission spectrometry (ICP-OES; Spectro Genesis; SPECTRO Analytical Instruments, Kleve, Germany). These were K, P, calcium (Ca), magnesium (Mg), manganese (Mn), iron (Fe), zinc (Zn), boron (B), and copper (Cu). Several quotients were calculated to more fully understand macronutrient pairs. These were N:P, N:K, and K:P.

2.2. Serianthes grandiflora and Serianthes nelsonii

A Philippine study was conducted in Lavazares, Samar from 14 November 2018 until 27 February 2019. The study site was on the floor of a copra farm positioned on a sand berm at 8 m asl. This study site was within in situ S. grandiflora habitat, and exact location is not included for conservation reasons as the site is within a Government Protected Area. The source tree which provided the S. grandiflora seeds was 326 m away from the study site. In addition to this native Serianthes species, the study also included Rota-sourced S. nelsonii seedlings. The methods were the same as for the S. kanehirae study, except the container substrate was the sand from the study site. The second dose of N and K fertilizer was applied on 28 December 2018. Incident light was determined by a hemispherical photograph from the center of the experimental layout and taken at 2 m elevation. The percent open sky was calculated digitally as 42%.

The layout was a 2 × 4 factorial in a completely randomized design with six replications. The beginning plant size for S. grandiflora was 27 ± 1 cm tall, 5.1 ± 0.3 mm stem diameter, and with 10 ± 1 leaves. The beginning plant size for S. nelsonii was 25 ± 1 cm tall, 4.8 ± 0.2 mm stem diameter, and with 8 ± 1 leaves. Growth in height and stem diameter were calculated from beginning and ending measurements in the 15-week study. Leaf elemental concentrations were measured and stoichiometry variables were calculated as previously described.

2.3. Statistical Analysis

All response variables were subjected to analysis of variance. The stoichiometry variables were log-transformed prior to analysis to meet parametric prerequisites. Some of the S. kanehirae variables exhibited unequal variances, so the PROC MIXED procedure (SAS Institute, Cary, NC, USA) was employed to conduct a one-way ANOVA. The variables for S. grandiflora and S. nelsonii met all parametric requirements, and the PROC GLM procedure was employed to conduct a two-way ANOVA with species and fertilizer treatments serving as the two factors. Treatment means were compared with Tukey’s HSD.

2.4. Soil Nutrients

The content of C, N, P, and K in the two substrates used as container media were determined from four random samples collected from the bulk substrate prior to separating into the nutrient treatments. Dry combustion was used to quantify C and N. Available P was determined by the Olsen method [18], and available K was quantified by ICP-OES following digestion with diethylenetriaminepentaacetic acid.

3. Results

3.1. Serianthes kanehirae

The substrate contained 82 ± 6 mg·g⁻¹ C, 12 ± 1 mg·g⁻¹ N, 29 ± 2 µg·g⁻¹ P, and 56 ± 3 µg·g⁻¹ K. Plant growth was stimulated by N additions, but not P or K additions for S. kanehirae plants grown for three months. Nitrogen fertilizer increased height growth 79% (p < 0.001), stem diameter growth 49% (p < 0.001), and leaf number 71% (p = 0.011) above that of the control plants. These plant traits for plants receiving P or K fertilizer did not differ from those of control plants (Figure 1).

The influence of the single nutrient fertilizer treatment on leaf tissue nutrient concentration was heterogeneous among the nutrients (

Table 1. The influence of N, P, or K fertilization on nutrient concentrations of Serianthes kanehirae leaves. C = control; P = phosphorus fertilization; K = potassium fertilization; N = nitrogen fertilization. Mean ± SE, n = 6.

Nutrient	C	P	K	N	p
Calcium 2	12.6 ± 0.6a 1	11.1 ± 0.9ab	9.5 ± 0.6b	8.1 ± 0.6c	<0.001
Carbon 2	415 ± 4	414 ± 5	413 ± 4	421 ± 4	0.091
Magnesium 2	2.7 ± 0.1b	3.1 ± 0.2a	2.4 ± 0.2bc	2.2 ± 0.2c	0.011
Nitrogen 2	17.2 ± 0.5c	21.2 ± 0.6b	19.4 ± 0.6bc	26.6 ± 0.7a	<0.001
Phosphorus 2	1.9 ± 0.1b	2.4 ± 0.2a	1.6 ± 0.1b	1.0 ± 0.1c	<0.001
Potassium 2	10.4 ± 0.4b	10.2 ± 0.3b	15.3 ± 0.4a	8.7 ± 0.2c	<0.001
Boron 3	36.7 ± 1.1a	38.0 ± 2.1a	29.7 ± 1.1b	24.9 ± 1.1c	<0.001
Copper 3	1.7 ± 0.1b	2.3 ± 0.2a	1.4 ± 0.1c	1.6 ± 0.2bc	0.008
Iron 3	61.3 ± 3.9a	59.3 ± 3.2a	45.3 ± 2.9b	36.8 ± 1.9c	<0.001
Manganese 3	51.9 ± 2.5a	37.8 ± 2.4b	39.8 ± 2.1b	30.8 ± 1.4c	<0.001
Zinc 3	33.2 ± 1.5a	26.4 ± 1.1b	22.1 ± 1.1c	36.2 ± 1.9a	0.008

¹ Means with same letter within each row are not different. ² (mg·g⁻¹) ³ (µg·g⁻¹).

). The only element that was not influenced by the fertilizer treatments was C. As expected, N fertilizer increased N concentration 55%, P fertilizer increased P concentration 26%, and K fertilizer increased K concentration 47% above control plants. Moreover, N fertilizer reduced the concentration of Ca, Mg, P, K, B, Cu, Fe, and Mn in comparison to the control plants. The influence of P fertilizer on concentrations of the other nutrients was varied. P fertilizer increased Mg, N, and Cu concentrations; and reduced Mn and Zn concentrations compared to control plants. The K fertilization treatment reduced Ca, P, B, Cu, Fe, Mn, and Zn concentrations compared to control plants.

The influence of single-element fertilizers on leaf stoichiometric variables followed expected patterns based on the direct influences on leaf N, P, or K concentrations. For example, fertilizing with N directly influenced each variable with N concentration as the numerator or denominator, and the same occurred for P or K fertilizers (

Table 2. The influence of N, P, or K fertilization on stoichiometry of nutrients for Serianthes kanehirae leaves. C = control; P = phosphorus fertilization; K = potassium fertilization; N = nitrogen fertilization. Mean ± SE, n = 6.

Variable	C	P	K	N	p
N:P	9.3 ± 0.4c 1	9.0 ± 0.4d	12.6 ± 1.4b	27.0 ± 1.9a	<0.001
N:K	1.7 ± 0.1c	2.1 ± 0.1b	1.3 ± 0.1d	3.1 ± 0.2a	<0.001
K:P	5.6 ± 0.3b	4.3 ± 0.2c	9.9 ± 0.8a	8.8 ± 0.6a	<0.001

¹ Means with same letter within each row are not different.

). However, the fertilizer treatments also influenced the stoichiometric variables that did not include the element in each of the fertilizers. For example, N fertilization greatly increased K:P; P fertilization increased N:K; and K fertilization increased N:P. Of the three fertilizer treatments, N fertilizer elicited the greatest relative changes in the values, because leaf N concentration was so elevated and leaf P and K concentration was so depressed with this treatment. Fertilizing with N increased the severity of P and K deficiencies as determined by the quotients; and fertilizing with K increased the P deficiency as determined by the quotients.

3.2. Serianthes grandiflora and Serianthes nelsonii

The sand substrate that served as the container substrate for the Philippine study contained 102 ± 8 mg·g⁻¹ C, 4 ± 1 mg·g⁻¹ N, 11 ± 1 µg·g⁻¹ P, and 39 ± 3 µg·g⁻¹ K. The growth traits were influenced by the factorial experimental approach differently, with plant height exhibiting a significant interaction between the two factors (p = 0.006), but the stem diameter (p = 0.417) and leaf number (p = 0.302) exhibiting no interaction. The interaction for height growth occurred because the S. grandiflora plants exhibited much more growth than S. nelsonii plants, not because the fertilizer treatments generated different patterns (Figure 2). Indeed, height increment of S. grandiflora plants was more than double that of S. nelsonii plants. As with S. kanehirae in the Guam study, N fertilizer greatly increased height growth but P and K fertilizer did not influence height growth.

The leaves which remained alive at the end of the study and the vertical position of the oldest leaves on the stems was used to estimate the oldest leaf age for each treatment. Control plant leaves were 9–10 weeks old when they died, plants receiving P fertilizer produced leaves that lived to about 12 weeks, plants receiving K fertilizer produced leaves that lived to about 14 weeks, and plants receiving N fertilizer increased leaf longevity to an estimated 17 weeks.

The increase in stem diameter was influenced independently by species (p < 0.001) and fertilizer treatment (p < 0.001). As with height, stem diameter growth was much greater for S. grandiflora (3.7 ± 0.3 mm, n = 24) than for S. nelsonii (2.9 ± 0.2 mm). The influence of fertilizer treatments on stem diameter growth followed patterns similar to the S. kanehirae study, in that N fertilizer roughly doubled stem diameter growth but plants receiving P or K fertilizer were not different from control plants (Figure 3a). The final leaf number was influenced independently by species (p < 0.001) and fertilizer treatment (p < 0.001). There were 15 ± 1 (n = 24) leaves for S. grandiflora and 9 ± 1 leaves for S. nelsonii. Leaf number was 10–12 per plant for control, P, and K treatments, which were significantly less than leaf number per plant for the N treatment (Figure 3b).

The leaf nutrient concentrations were influenced by the factorial experimental approach differently, with N concentration (p = 0.006) and P concentration (p < 0.001) being the only nutrients that exhibited an interaction between the two experimental factors. The different responses between the species appeared to be caused by N and P fertilizer treatments, as the influence of K fertilizer was similar for the two species (

Table 3. The influence of N, P, or K fertilization on nitrogen and phosphorus concentrations (mg·g⁻¹) of Serianthes nelsonii and Serianthes grandiflora leaves. C = control; P = phosphorus fertilization; K = potassium fertilization; N = nitrogen fertilization. Mean ± SE, n = 6.

Nutrient	C	P	K	N	p
		S. nelsonii
Nitrogen	13.2 ± 0.3d 1	17.7 ± 0.5b	15.1 ± 0.5c	23.1 ± 0.6a	<0.001
Phosphorus	2.0 ± 0.1b	2.6 ± 0.1a	1.8 ± 0.1b	1.2 ± 0.1c	<0.001
		S. grandiflora
Nitrogen	16.8 ± 0.4c	19.9 ± 0.6b	18.8 ± 0.5b	27.5 ± 0.6a	<0.001
Phosphorus	1.5 ± 0.1b	2.4 ± 0.1a	1.5 ± 0.1b	1.1 ± 0.1c	<0.001

¹ Means with same letter within each row are not different.

). For example, N fertilization increased N concentration 75% above the control plants for S. nelsonii but only 63% for S. grandiflora. Similarly, P fertilization increased P concentration 30% above the control plants for S. nelsonii and 60% for S. grandiflora. As with the Guam study, N fertilization decreased P concentration below that of control plants for both species.

The foliar nutrients other than N and P were not influenced by the factorial interaction, but they were influenced by the single-element fertilizer treatments. Therefore, the leaves of both species behaved similarly in response to the edaphic nutrient manipulations. These S. grandiflora and S. nelsonii plants responded to the fertilizers with patterns that were similar to the S. kanehirae study (

Table 4. The influence of N, P, or K fertilization on nutrient concentrations of Serianthes leaves. Mean of Serianthes grandiflora and Serianthes nelsonii. C = control; P = phosphorus fertilization; K = potassium fertilization; N = nitrogen fertilization. Mean ± SE, n = 12.

Nutrient	C	P	K	N	p
Calcium 2	13.0 ± 6a 1	12.5 ± 0.6ab	11.2 ± 0.6b	8.7 ± 0.4c	<0.001
Carbon 2	419 ± 2b	418 ± 3b	417 ± 2b	425 ± 3a	<0.001
Magnesium 2	2.8 ± 0.1ab	3.1 ± 0.2a	2.6 ± 0.2b	2.1 ± 0.2c	<0.001
Potassium 2	11.0 ± 0.4b	11.2 ± 0.4b	15.5 ± 0.4a	8.8 ± 0.2c	<0.001
Boron 3	36.8 ± 1.9a	38.2 ± 2.2a	32.1 ± 1.7b	29.5 ± 1.7c	<0.001
Copper 3	1.9 ± 0.2ab	2.3 ± 0.2a	1.8 ± 0.2b	1.9 ± 0.2ab	<0.001
Iron 3	63.5 ± 3.5a	60.9 ± 2.3a	47.6 ± 2.6b	39.8 ± 1.6c	<0.001
Manganese 3	51.9 ± 2.2a	38.0 ± 1.9b	42.2 ± 1.3b	30.5 ± 1.2c	<0.001
Zinc 3	34.1 ± 2.1a	30.0 ± 1.8b	27.3 ± 2.1b	37.8 ± 2.8a	<0.001

¹ Means with same letter within each row are not different. ² (mg·g⁻¹) ³ (µg·g⁻¹).

). Fertilizing with N reduced the concentration of Ca, Mg, K, B, Fe, and Mn in comparison to the control plants. Fertilizing with P reduced only Mn and Zn concentrations, and fertilizing with K reduced Ca, B, Fe, Mn, and Zn concentrations compared to control plants.

The influence of single-element fertilizers on leaf stoichiometric variables followed patterns that were similar to the S. kanehirae study (

Table 5. The influence of N, P, or K fertilization on stoichiometry of nutrients for Serianthes leaves. Mean of Serianthes grandiflora and Serianthes nelsonii. C = control; P = phosphorus fertilization; K = potassium fertilization; N = nitrogen fertilization. Mean ± SE, n = 12.

Variable	C	P	K	N	p
N:P	9.0 ± 0.8b 1	7.6 ± 0.6c	10.8 ± 1.1b	23.6 ± 1.4a	<0.001
N:K	1.4 ± 0.1c	1.7 ± 0.1b	1.1 ± 0.1d	2.9 ± 0.2a	<0.001
K:P	6.3 ± 0.3c	4.5 ± 0.2d	9.8 ± 0.5a	8.1 ± 0.2b	<0.001

¹ Means with same letter within each row are not different.

). Fertilizing with N greatly increased N:P, N:K, and K:P for these two Serianthes species. Fertilizing with P decreased N:P, and K:P; and increased N:K. Fertilizing with K decreased N:K and increased K:P.

4. Discussion

Single nutrient fertilization studies have provided valuable information concerning limitations of plant growth in tropical and subtropical forests, e.g., [17,19,20]. My study showed that single-nutrient fertilization of Serianthes plants imposed by adding N or K increased foliar concentration of each nutrient, but also generated numerous changes in concentration of other nutrients. For N, this was partly by way of the dilution effect associated with growth stimulation. In contrast, adding P fertilizer to the soils increased foliar P concentration without substantial changes in concentration of other nutrients. For example, the addition of P fertilizer did not reduce foliar concentrations of N or K below that of control plants. In terms of N:P:K relations, these findings indicate that fertilizing Serianthes plants with N or K may compromise plant health by magnifying P deficiencies. In contrast, fertilizing Serianthes plants with P may be an effective conservation method for improving plant nutrition without influencing N or K deficiencies. This new knowledge may be put to use for managing in situ S. nelsonii plants, because P is the most limiting nutrient based on leaf elemental stoichiometry [15].

An unexpected outcome from the Philippine study was that S. grandiflora plant growth greatly exceeded S. nelsonii plant growth. Previous studies in Guam with these species generated plant growth that was similar for the two species [1,2]. There are two possible explanations for these findings. First, the genetic differences in the seed sources may have elicited disparate vigor among the studies. The Guam studies were conducted with S. grandiflora seeds from a Bohol provenance, and the current Philippine study was conducted with seeds from a northern Samar provenance. Moreover, the previous studies were conducted with S. nelsonii seeds from Guam, and this Philippine study was conducted with seeds from Rota. Second, this may be an example of home field advantage. The Guam nursery studies could be viewed as off-site studies for all three species. The S. nelsonii plants in the Philippine study were also off-site. However, the source tree for the S. grandiflora plants in this study was only 326 m away from the study site. Moreover, the study was conducted with edaphic substrates within the environment of in situ habitat. More research is needed using fertilizer manipulation treatments with in situ S. nelsonii trees on Guam and in situ S. kanehirae trees in Yap or Palau to more fully understand home field advantage in relation to nutrient budget at the whole-plant scale.

Leaf longevity was not directly measured, but the ending leaf number in relation to the beginning leaf number was used to indicate that the N fertilizers greatly increased leaf longevity. Fertilization with N has been shown to delay fall senescence of deciduous tree leaves in temperate regions [21]. Experimental approaches that directly quantify the influence of plant nutrition on Serianthes leaf longevity are clearly needed.

Various federal agencies are currently coordinating efforts to improve Guam forest health and resilience by increasing native plant coverage [22]. This new initiative augments the ongoing efforts to recover threatened and endangered plant populations. Serianthes nelsonii is critically endangered [5,6], and is among the palette of plant species that are actively being planted to address the goals of the initiative. A full understanding of the plant nutrition needs and the contributions of this Fabaceae tree to ecosystem nutrient cycling will be required to make appropriate management decisions. This handsome endemic tree produces leaf litter with nutrient content and stoichiometry that predicted rapid litter decomposition [15]. This prediction was shown to be true by quantifying carbon dioxide efflux and nitrogen release from litter incubated in native soil [23]. A complete nutrient solution may be sprayed on S. nelsonii foliage to improve plant nutrition [16]. The current study expands our understanding of this subject by revealing the utilitarian benefits of P-only fertilizers for mitigating P deficiency.

Fertilizing S. nelsonii plants with N or K should be restricted to fertilization programs with balanced nutrient fertilizers, ensuring the addition of all macronutrients to counteract the imbalances caused by single-element fertilizers. In contrast, fertilizing with P alone is a viable approach for mitigating P deficiency and improving plant nutritional status. Unfortunately, the federal permitting regulations on Guam do not allow the use of field soil in S. nelsonii nursery research, as container media components must rely on imported foreign materials. Therefore, employing the methods in this study within a S. nelsonii conservation nursery in Guam is not possible unless these regulations are relaxed. This preliminary finding that P-only fertilization can increase leaf P content without causing a substantial dilution effect or imbalance of other nutrients justifies the implementation of long-term replicated in situ fertilization studies to confirm these findings at the habitat scale and to inform ecological stoichiometry questions.

5. Conclusions

The growth and nutritional responses of plants to single nutrient fertilizer applications may inform horticultural and conservation management decisions. Little is known about nutrient management for Serianthes species. Fertilizing three Serianthes species with N-, P-, or K-only fertilizer indicated a P-only fertilizer may be effective in mitigating P deficiencies without causing imbalances and complications with other nutrients. Conservation practitioners may use this approach to correct for P deficiency within the endemic range of S. nelsonii where P deficiency has been observed.