Coconut Coir as a Sustainable Nursery Growing Media for Seedling Production of the Ecologically Diverse Quercus Species

Peat, a non-sustainable resource, is still predominately used in forest nurseries. Coconut coir might provide an alternative, renewable, and reliable growing media but few studies have evaluated this media type in forest nurseries. We assessed the influence of pure coir, in combination with various fertilization regimes, on the growth and physiology of three ecologically diverse Quercus species seedlings (Q. robur, Q. pubescens, and Q. ilex) during nursery cultivation. Seedlings were grown using peat and pure coir in combination with three fertilization treatments (standard, K-enriched, and P-enriched). Data were collected for: (1) growth and physiological traits; (2) detailed aboveand below-ground morphological traits by destructive analysis; and (3) NPK content in leaves, shoot and roots, and in the growing media, following cultivation. Peat and coir in combination with the various fertilization treatments affected aboveand below-ground morphology and, to a lesser extent, the physiological traits of Quercus seedlings. Large effects of the substrate occurred for most morphological variables, with peat being more effective than coir in all studied species. Fertilization also produced significant differences. The effect of K-enriched fertilization on plant growth was clear across the three species and the two growing media. P-enriched fertilization in peat was the only combination that promoted a higher amount of this element in the tissues at the end of cultivation. Despite their smaller size, seedlings produced in coir were compatible with standard Quercus forest stocktype size, and showed a proportionally higher root system development and fibrosity. Our results suggest that coir can be used as an alternative substrate to grow Quercus species seedlings, and that fertilization can offset coir deficiencies in chemical properties. As several functional traits drive planting performance under varying environmental conditions. according to the Target Plant Concept, coir might thus serve as an acceptable material for seedling cultivation in some cases.


Introduction
Forest seedling field performance is related to plant morphological and physiological characteristics [1][2][3]. Development of morpho-physiological traits, in turn, is strongly influenced by nursery practices [1,[4][5][6][7]. In container plant production, the effectiveness of the growing media is information is available for the forest nursery sector (i.e., [11,61,62]). Moreover, coconut fiber is often tested as a component at different proportions of a soilless growing media mixture with other materials, complicating assessment of coir as a single component [9], especially in relation to deficiencies in its chemical properties (i.e., CEC). In this sense, the use of controlled release fertilizer (CRF) with coconut fiber could help to offset the mentioned low CEC.
In this study, we focused on the Quercus species that is widely used in the Mediterranean region for reforestation, afforestation, and forest restoration projects [63][64][65][66]. The species were chosen according to different ecological adaptations, particularly in relation to Mediterranean or arid climates [67]-from the most well adapted to a Mediterranean climate with dry summers, Q. ilex (holm oak), to the intermediate Q. pubescens (pubescent or downy oak), to the less Mediterranean adapted, Q. robur (pedunculate oak), which prefers a temperate climate without a dry season, is widespread across Europe, including the Mediterranean region, and is one of most common species for multi-purpose hardwood forest plantings in Europe [68,69]. We aimed to improve the knowledge of the influence of pure coir on the growth and physiology of these species during nursery cultivation. Coir was combined with three different fertilization treatments (nursery standard, enriched in K, enriched in P) to assess the combined effects on seedlings in relation to-height, morphological traits, physiological traits, and NPK content both in the shoot-and root-system. Specifically, we addressed the following questions: (1) Is pure coir effective in growing oak seedlings to be used in forest plantation projects? (2) Which traits are mainly influenced by this renewable and more sustainable growing media? (3) Is it possible to offset coir deficiencies in chemical properties by fertilization? (4) Do the studied species, which have different leaf habits despite different ecological adaptations, respond differently to fertilization and substrate treatments?

Nursery Stock Cultivation
The nursery stock was grown in 2017 in a central Italy nursery (43 •  Peat substrate (Pe), considered as control, was made of 70% coarse Baltic peat moss (0-40 mm particles size) and 30% pumice (5-8 mm), which was added to avoid excessive compaction of pure peat. Main traits of the mixture were-pH 5.7, EC 0.43 dS/m; bulk density 298 kg/m 3 , and porosity 86.9%. Coir (Co) was composed of 30% coconut fiber (fibrous material that constitutes the thick mesocarp of the nut) and 70% coconut pith (fine residual material), with pH 5.2, EC 0.38 dS/m; bulk density 122 kg/m 3 , and porosity 92.5%. The week before sowing, the cavities were uniformly filled with growing media to 0.5 cm beneath the top.
Standard fertilization (St), a Controlled-Release fertilizer (CRF) fertilizer widely used in Italy to commercially grow oak seedling nursery stock, was considered as control, and it was compared to two alternative commercial fertilization formulas-enriched in potassium (K-enriched) or phosphorus (P-enriched). The St formula was Osmocote Exact Standard 12-14 months NPK 15-9-11 plus micronutrients at 3 kg·m −3 . P-enriched substrate contained an addition of 19% P 2 O 5 fast release (single superphosphate, SSP, 2 kg·m −3 ) to control, while K-enriched substrate was obtained by 4 kg·m −3 of Osmocote Exact Standard High K 12-14 months NPK 11-11-18 plus micronutrients. N, P, and K amount per seedling is shown in Table 1.  50.0 E). The 1000-seed weight and germination of the three seed lots (according to International Seed Testing Association procedures [70]) were 6.0 kg and 80% for Q. robur, 2.4 kg and 68% for Q. pubescens, 3.2 kg and 83% for Q. ilex, respectively. From October 2016 to April 2017, the acorns were stored at 3 ± 0.5 • C in moist sand to simulate normal overwintering and to prevent acorn germination before the experiment started. In March 2017, the seeds were moved outside to stimulate germination, and then the pre-germinated acorns (0.5 cm maximum radicle length) were sown in multi-pots placed under a tunnel protected by a transparent plastic film that was removed at mid-May. Seedlings were irrigated daily by sprinklers following seedling evapotranspirative demands (i.e., spring: 5 L·m −2 in 6 min; summer and autumn: 40 L/m 2 and 20 L/m 2 , respectively, in 24 min). Environmental conditions (air temperature and humidity) were monitored by a weather station (inside the tunnel) and substrate moisture was assessed once a week (6 pots per stocktype, Soil moisture meter PCESMM1, PCE Instr. Corp., PCE Holding, Hamburg, Germany).

Data Collection
Emergence and height were measured weekly from the end of March until September. Here, final height data are presented. During the season, physiological traits, such as chlorophyll content (SPAD-502, Konica-Minolta Sensing Europe B.V.) and chlorophyll fluorescence (ChlF, by Handy PEA, Hansatech, UK) were measured monthly on a sample of three fully expanded leaves on 9 seedlings per stocktype in dark-adapted leaves (for at least 40 min), during the growing season (June, July, August). The PSII functionality was described by the F V /F M (ratio of Variable to Maximum Fluorescence) to Strasser et al. [71]. Results related to physiological traits were shown only for August, the date closer to plant lifting, in supplementary material.
Macro-element (N, P, K) were assessed in October (before leaf abscission) on 12 seedlings per stocktype (S × F) per species, merging 2 plants of the same multi-pot, by Nutrilab (Universidad Rey Juan Carlos, Madrid, Spain). Briefly, analysis of leaf, stem, and root N and P concentration of these samples was done by the standard Kjeldahl method, while K concentration was determined using a perchloric acid extraction. Substrate analysis at the end of the nursery cultivation was performed by Denetra snc (Pescia, Italy) on 9 seedlings per stocktype (S × F) per species. Concurrently, seedling morphological traits were assessed, using 20 destructively sampled seedlings per treatment combination (360 in total). Height (H), number and dry biomass of leaves (Ln and Lb), root-collar diameter (RcD), and dry biomass of the shoot (stem) were measured. Root-system was assessed by main root (tap-root) dry biomass (MRb); first-order lateral roots (FOLR) dry biomass (FOLRb) for three diameter classes (<1 mm, 1-5 mm, >5 mm) measured at the junction with the tap root, root-system volume (by immersion), and dry weight and density (Rv, Rb, and Rd, respectively). Shoot to root ratio (S/R), H/RcD, and specific leaf area (SLA) were also calculated.

Statistical Analysis
A randomized complete block design (3 blocks) was used. Each block included the 6 randomized S × F combinations per species; each combination comprised 48 sowed cavities (over 4 multi-pots), for a total of 2592 sowed containers. In a preliminary multifactorial ANOVA (model, Y ijlmn = µ + Block i + Date j + Substrate (S) k + fertilization(F) l + S × F interaction kl + error ijklm ), the block effect was not significant for any morphological variables, highlighting homogeneous growing conditions. Thus, we removed block as source of variation and a multifactorial ANOVA was performed, separately for each species to avoid complex higher-level interactions, considering substrate (S), fertilization (F), and their interactions (S × F) as a source of variation. In case of significant results (p ≤ 0.05), the Tukey post-hoc test was used for multiple comparisons (α = 0.05) to highlight homogenous groups within species. In this study, we present results related to substrate (S), fertilization (F), and their significant interactions (S × F). To provide information about the traits of the studied stocktypes (combinations S × F), Tukey test results among combinations are shown in supplementary material. StatSoft Statistica 11 (Tulsa, OK, USA) was used to process all data.

Emergence, Growth, and Physiological Traits during Nursery Cultivation
Final emergence was not affected by either factor in the three studied oaks. In the Q. ilex, seedlings emergence started between 3 and 4 weeks later than in the other two species. In all species, substrate affected height beginning relatively early, from June (from May in Q. robur, data not shown), and seedlings in Pe were higher than that in Co (Table 2); in all species, fertilization occurred later (September in Q. robur, July in Q. pubescens, July in Q. ilex), and K-enriched fertilization promoted taller seedlings ( Table 2). As a result, at the end of the growing season, the tallest stocktype was grown in Pe-K in all species, with differences among the stocktypes in coir, and generally, seedlings grown in Co-St performed worse (Table S1). Interaction was not significant in any case, excluding Q. robur in August and Q. ilex in August and September.
In general, in all species, in August ( Figure S1), both substrate (Pe > Co) and fertilization (K generally higher) affected the ChlF content, generally with no interactions (data not shown). In Q. robur and Q. pubescens, seedlings in peat showed higher values of F V /F M than in coir (significant in Q. robur, Figure S1), and K-enriched fertilization generally resulted in higher F V /F M values (significant in Q. ilex).

Morphological Traits
In Q. robur, both substrate and fertilization affected most of the analysed variables without any interaction (Tables 2 and 3), and generally, Pe and K-enriched fertilization were more effective in promoting seedling growth. For total dry biomass, the differences were as follows: Pe > Co by +22.6%, K > P and St by +13.4% and +36.6%, respectively ( Figure 1). Peat also promoted shoot-system and leaf biomass, SLA, as well as FOLR1-5 and total biomass of roots. K-enriched fertilization positively influenced shoot development, SLA, FOLR1-5, root system, and leaf biomass ( Figure 1 and Figure  S2). Pe stimulated biomass accumulation more in the shoot-than in the root-system with S/R values > 0.5 (Table 3), and K-enriched fertilization promoted higher S/R than St. Root volume was affected by substrate (Pe > Co) and fertilization (K > St); root density was influenced only by fertilization (Co-St > Pe-K). Table 2. Multifactorial ANOVA and Tukey post-hoc test results (p ≤ 0.05 in bold) for seedling morphological traits (mean ± SD) at the end of the season (N obs = 20 seedlings per stocktype per species). Source of variation-substrate (S), fertilization (F), and their interaction (S × F). Capital and lowercase letters indicate homogeneous groups for S and F, respectively. Variables: H (height); RcD (root collar diameter in mm); H/RcD (seedling taper); Rv (root-system volume in cm 3 ); Rd (root-system density = Rb/Rv in g cm −3 ); S/R (shoot-root ratio); SLA (specific leaf area, in cm 2 g −1 ); Ln (number of leaves); and Lb (leaf dry biomass in g). In Q. pubescens, substrate affected morphological variables more than fertilization, and, generally, in the case of significant differences, seedlings in peat and K-enriched fertilization were better developed (Tables 2 and 3, Figure 1). Total biomass in Pe was higher than in Co by +78.3%, and the gap between K and St and P was +19.3% and +20.1%, respectively. Pe significantly promoted all biomass variables excluding FOLR > 5. Fertilization with K promoted biomass growth in the shoot-system, leaves, and in FOLR1-5 over St fertilization (Figure 1 and Figure S1). Shoot-to-root ratio was lower than 0.51 in all stocktypes (Table 2) and was affected by both substrate (Pe > Co) and fertilization (K > P and St). Root volume of Q. pubescens seedlings raised with peat was higher than that of coir, while the opposite occurred for root density.
As per previous species, in the case of Q. ilex, the substrate affected the majority of the analyzed variables (with a Pe > Co pattern), and generally, when fertilization was significant, seedlings grown in K-enriched fertilizer had higher values (Tables 2 and 3, Figure 1). Total biomass was affected by both factors (Pe +63.5% than Co; K +36.2% and +50.2% than P and St, respectively). Similar differences (Pe > Co) occurred for shoot-system and leaf biomass, main root, FOLR < 1, FOLR1-5, and root biomass. Fertilization affected leaf, main root, FOLR1-5, and root biomass (K> P and St). S/R ratio was affected only by substrate (Table 3), both in Co and Pe it resulted higher than 0.50. Root volume was affected by both substrate (Pe > Co) and fertilization (K > P and St).
(Pe > Co) occurred for shoot-system and leaf biomass, main root, FOLR < 1, FOLR1-5, and root biomass. Fertilization affected leaf, main root, FOLR1-5, and root biomass (K> P and St). S/R ratio was affected only by substrate (Table 3), both in Co and Pe it resulted higher than 0.50. Root volume was affected by both substrate (Pe > Co) and fertilization (K > P and St).

N, P, K Concentration in Seedlings and Substrate at the End of Cultivation
In Q. robur, N concentration was not affected by treatments, excluding fertilization on root-system (St and P > K), while both substrate (Pe > Co) and fertilization (P and K > St) affected P concentration in seedling parts (excluding substrate on shoot; Table 4, Figure 2). Pe-P had higher P concentration than other stocktypes in leaves and in shoot ( Figure S3). Both S and F did not influence K concentration. In Q. pubescens, N concentration in any tissue was not influenced by either treatment (Table 4, Figure 2); P concentration was affected by both substrate (Pe > Co) and fertilization (P and K > St; excluding substrate on shoot-system Table 4, Figure 2). Pe-P had a generally higher P content than other stocktypes in leaves, shoot, and roots ( Figure S3). Differences in K concentration were found between substrates in leaves (Pe > Co). In Q. ilex, both substrate (Co > Pe) and fertilization (St > K) influenced plant N concentration in all tissues (Table 4, Figure 2). Substrate affected leaf P concentration (Co > Pe) and fertilization affected the root-system concentration (P > St), K concentration was influenced by substrate in roots (Co > Pe), and by fertilization in the shoot-system (Table 4, Figure 2). Nitrate N concentration of the growing media with Q. robur was affected by substrate and fertilization (Pe > Co; P > K), and only by fertilization for ammonium N (P > K, Table 5). In Q. pubescens, fertilization affected P concentration (P > K), and both nitric and ammonium N (P > K). In Q. ilex, nitrate N was affected by fertilization (P > K) and ammonium N by both factors (Co > Pe; P > K), and the P concentration result was not affected by substrate and fertilization, while both affected K concentration (Co > Pe; K > P and St). No interaction among factors occurred.  Table 4. p values of multifactorial ANOVA (p ≤ 0.05 in bold) for macro-element concentration (N, P, K) in leaves, shoot-system, and root-system, at the end of the growing season (N = 6 couples per stocktype). Source of variation-substrate (S), fertilization (F), and their interaction (S × F). Nitrate N concentration of the growing media with Q. robur was affected by substrate and fertilization (Pe > Co; P > K), and only by fertilization for ammonium N (P > K, Table 5). In Q. pubescens, fertilization affected P concentration (P > K), and both nitric and ammonium N (P > K). In Q. ilex, nitrate N was affected by fertilization (P > K) and ammonium N by both factors (Co > Pe; P > K), and the P concentration result was not affected by substrate and fertilization, while both affected K concentration (Co > Pe; K > P and St). No interaction among factors occurred.

Discussion
Our study highlighted the early effects of substrate on growth (2 months after sowing in Q. robur and Q. pubescens; 3 months in Q. ilex, whose seedlings emerged later) and persisted through cultivation. The absence of any fertilization effect on early growth phases was expected-dependence of Quercus seedlings on acorn nutrients decreased as the seedlings developed [7,72]; thus, first growth flush in many Quercus species was to a higher extent related to acorn size and nutrients [73,74], rather than to the quality of the growing media [7,72], and our results were in line with those of Villar-Salvador et al. [72] in Q. ilex. At the end of cultivation, seedlings of all species grown in peat were taller than those grown in coir. However, the height of seedlings in coir were at least equal to that of comparable forest nursery production systems of the Quercus species [75,76]; in other cases height was higher both in Q. robur [77,78] and in Q. ilex [27,40,79,80], and were in line with the Italian national regulation [81].
Similar trends were found in other studies with other species using substrate mixtures that included coir-Rose and Haase [61] with Douglas fir, Tsakaldimi and Ganatsas [11] with, among others, Q. ilex, and Radjagukguk et al. [62] with two Eucalyptus. Offord et al. [82] did not observe differences in several species, including Eucalyptus melliodora. The reduced growth rate observed in seedlings grown in coir could be due to the lower CEC than in peat, which reduces nutrients availability for seedlings [10]. This could explain the smaller size of the vast majority of the studied morphological traits in all coir stocktypes in all species, regardless of fertilization. Additionally, according to previous studies, such as Handreck [83] and Grantzau [84], there is a greater immobilization of soluble nitrogen in coir than peat suggesting the need to add extra N fertilization during cultivation. In our study we did not find a lower tissue concentration of seedlings raised with coir, suggesting an absence of nitrogen deficiency. The common mixtures and fertilization protocols used by nursery companies in horti-flori-fructiculture sector have been optimized [85] and plants in coir grow equally as peat mixtures [29,82,86,87], suggesting that fertilizer can offset substrate deficiencies.
According to our results, morphological traits variability was explained more by substrate than by fertilization. Fertilization effects on growth occurred later than those of the growing media, in mid-summer for Q. pubescens and Q. ilex, and near the end of the growing season in Q. robur. In all species and in both growing media, the K-enriched fertilization improved the growth rate during the growing season and promoted taller plants. Apart from N fertilization, scarce literature is available on the effect of the other macro-elements on the growth of forest species in the nursery [7]. K is recognized as a key element of many metabolic processes [88], some of which are related to plant stress responses. Direct connections of K with seedling growth are less investigated, but in a meta-analysis of forest species (including plants at seedling stage), Tripler et al. [89] highlighted that plant growth responded positively to an increase in K availability. Studies on tropical forest species highlighted a direct effect of K fertilization in promoting plant growth with a higher shoot/root ratio [90,91]. Similar to height responses, seedlings in all species grown in peat and K-enriched fertilization allocated more resources to above-ground dry biomass (leaves and shoot system). Similar results for substrate effects were obtained by Rose and Haase [61] in Douglas-fir. Coir K content is usually higher than that in peat [31,61]. However, in our study, despite the tested rates of K being high compared to the literature on oaks [92][93][94], we observed an outstanding effect of such fertilization on plant growth regardless of species or substrate, suggesting K deficiency status. K-enriched fertilization included an additional amount of P, as compared to a standard fertilizer, which could have been helpful in sustaining seedling growth. However, fertilization with P was not as effective as K in promoting height and biomass in the studied species, although, generally, it was better than standard fertilization. P plays a key role in plant metabolism, being a critical element for many physiological reactions [95] and root morphological traits (structure, growth, and articulation) [49,50,[95][96][97]. Thus, a non-pronounced effect on the shoot system growth was expected.
In all studied species and treatments, peat and K-enriched fertilization promoted the greatest below-ground development. In all cases, shoot/root ratio was < 1, and cultivation in coir reduced this ratio further, particularly for Q. pubescens. Other studies on flowering species pointed out that cultivation in coir can promote a proportionally higher root system component [86,98]. The same effect occurred on the Douglas fir, with a higher root biomass for seedlings in coir and a decreasing trend in the S/R ratio of peat to a mixture of coir and peat, followed by that of peat to coir [61]. In contrast, Tsakaldimi and Ganastas [11] found a higher S/R ratio in 1-year-old seedlings of Q. macrolepis and Q. ilex grown in a mixture of peat and coconut fiber than in peat with perlite. On the other hand, Chulaka et al. [99] and Wilson et al. [100] reported the effects of coir-based substrates on S/R ratio with non-significant results versus peat-based substrates, and Colla et al. [98] found a higher S/R ratio in horticultural species grown in peat. A lower shoot-to-root ratio has been linked to survival under drought condition in holm oak [80], in Mediterranean shrubs [101], and savanna species [102]. On the contrary, Villar-Salvador et al. [40] highlighted that Q. ilex seedlings with a higher S/R had a lower mortality and a larger growth than those with the opposite attributes; however, we found quite similar shoot-to-root values for holm oak produced in coir to the highest values observed in this study (0.59 vs. 0.63, respectively).
Peat also promoted main root biomass and favored root fibrosity, especially the two smaller classes of FOLR (>1 mm and 1-5 mm). The proportion of FOLR on total root biomass was statistically affected by substrate (data not showed) in Q. robur and Q. ilex; however, the gap between substrates in all species was slight. This result suggests that plants grown in coir, despite a lower absolute root-system biomass, maintained comparable root fibrosity to peat. This trait, as well as a lower shoot-to-root ratio is crucial to resist water stress, and is related to post-planting survival and growth under arid conditions, such as in Mediterranean environments [103]. In seedlings grown with K-enriched fertilizer, the S/R ratio was higher but the biomass allocated to FOLR was positively influenced by such fertilization. However, studies on tropical tree species and in alpine and dryland environment [90,91] reported that K promoted shoot biomass and less FOLR, resulting in a higher S/R ratio. P-enriched fertilization did not promote the growth and articulation of the root-system, as expected, even though this study did not compare different levels of P and K, but rather three different fertilizations. Scarce literature has focused on the direct effect of P on detailed morphology below ground traits in Quercus. According to Pemán [7], the root systems of the Quercus species can be affected by the available P. In Q. ilex, Sardans et al. [104] found that P fertilization can promote root growth, Villar-Salvador et al. [40] linked P concentration to root regeneration, and Oliet et al. [47] highlighted a clear effect of root P concentration on root growth potential (RGP).
Physiological results are in agreement with morphology, demonstrating an overall better performance of photosynthetic machinery (considering SPAD and F V /F M in conjunction) of seedlings in peat and seedlings fertilized with enriched K, in most species. Chlorophyll fluorescence, even though measured values show absence of stress, has proven to be a sensitive technique to detect differences among stocktypes, especially those from different fertilization treatments. Results for Chlorophyll content were generally consistent with what was observed in morphology, and peat and K-enriched fertilization promoted higher values in all species. The values of chlorophyll content observed for deciduous oaks were in line with those in other studies [105,106].
We did not find a marked effect of substrate or fertilization on N concentration in plant tissues; at the end of cultivation. N concentration of Q. robur and Q. pubescens did not depend on the substrate and only sporadically depended on fertilization, whereas both factors affected Q. ilex. Despite a higher N content per seedling than that from other studies on the same species [92,107], higher than the sufficiency level and the optimum target in Q. ilex [92], N concentration in holm oak roots and the shoot-system was lower. We supplied 27% less N than that of Berger and Glatzer [108], as luxury consumption in Quercus petraea (a European temperate species), and it did not affect Q. robur tissue N concentration. However, levels of uptake and consumption can vary not only in relation to experimental factors, but also in terms of intrinsic ecological differences among provenances and species [92,94]. This makes the hypotheses on this issue for Q. pubescens uncertain, as it is not well-represented in the literature for this species.
Peat as well as P-enriched fertilization were effective in increasing the level of P in all plant tissues in Q. robur and Q. pubescens, while in Q. ilex, such fertilization was effective only for P concentration in roots, and in contrast to the other species, holm oak seedlings in coir performed better than in peat. Higher P root tissue concentration is important to promote root growth capacity after planting [109][110][111][112][113]. Campo et al. [80], in a study designed to define the quality standards for Q. ilex nursery stock, indicated that the values of N and P foliar concentration should be higher than 10 and 0.9 mg g −1 , respectively, to improve growth performance after planting. Considering this target, our results for N were higher and better in peat and in P-enriched and standard fertilization, and for P, these were about 10% lower and higher in peat. No references are available for Q. robur and Q. pubescens; however, in our study, foliar N and P concentrations in these deciduous species were higher than the standard values for the evergreen Q. ilex.
In our study, K concentration in plant tissues was generally not affected by both substrate and fertilization in all species, suggesting that K-enriched fertilization did not increase K concentration. Considering the higher performance of seedlings grown in K-enriched fertilization, regardless of species or substrate, this result strengthens the hypothesis that the Quercus species benefits from the high K availability and that the amount provided in the standard and in P-enriched fertilization was deficient. According to [93], K is the most responsive nutrient at deficiency status. The vast majority of literature on fertilization of forest nursery stock has focused on the N effects; thus, more investigations on the optimal K rates to maximize growth is needed. Del Campo et al. [80] concluded that nursery stock with higher K concentration can exhibit improved establishment success due to K retranslocation. Andivia et al. [114] observed that fertilization with K positively influenced morphological development and NPK content. K effects on field performance on Pinus halepensis have been observed with conflicting results [115,116]. Del Campo et al. [117] suggested that these multiple responses could be related to the interference of K with other nutrients, particularly nitrogen. The interaction between K fertilizers and different types of substrate requires further investigation.

Conclusions
Peat and coir in combination with different fertilization treatments affected above-and below-ground morphology and, to a lesser extent, physiological traits of Quercus seedlings in forest nursery production. The three studied species, despite being ecologically diverse, provided similar results, with peat and K-enriched fertilization resulting in larger seedlings and slightly improved physiological responses. Even though P-and, mostly, K-enriched fertilization partially offset the difference, seedlings in coir were smaller, which could be linked to deficiencies in the chemical properties of this material. Seedlings produced in coir showed a proportionally similar root system development and fibrosity, and a generally lower shoot-to-root ratio than seedlings grown in peat.
According to the target plant concept [3,5], functional traits driving planting performance vary according to environmental and operational conditions; and plants with a lower shoot-to-root ratio might perform better under water stress or in semiarid/arid environments [80,118]. Pure coir might thus serve as an acceptable material for seedling cultivation in such cases, despite the tendency to produce smaller seedlings than peat. However, a more detailed evaluation of the response of seedlings produced with coir to water stress is needed. NPK content was slightly influenced by treatments, although P-enriched fertilization in peat was the only combination that promoted a higher amount of this element in tissues at the end of cultivation; the strong effect of K-enriched fertilization on seedling growth and biomass suggested a K deficiency in the other fertilization treatments; however, macro-element content in relation to varying fertilization should be further investigated through a comparison of the single element rates.
Supplementary Materials: The following are available online at http://www.mdpi.com/1999-4907/11/5/522/s1. Table S1: Multifactorial ANOVA and Tukey post-hoc test results (p ≤ 0.05 in bold) for seedling morphological traits (mean ± SD) at the end of the season (N obs = 20 seedlings per stocktype). Figure S1: Multifactorial ANOVA and Tukey post-hoc test results of FV/FM values (box whisker plot) and SPAD units (mean and SD) analyzed for substrate and fertilization in August. Figure S2: Multifactorial ANOVA and Tukey post-hoc test results (p ≤ 0.05) for seedling dry biomass (g) allocation at the end of the growing season (N obs = 20 per stocktype per species). Figure S3: Macro-element concentration (mg g −1 for N. P. K) in the leaves, shoot-system, and root-system in stocktypes (mean ± SD).