Effect of Soil Mixtures Based on a Gneiss‑Derived Soil and Two Forest Floor Types on Growth and Nutritional Status of Castanea sativa Mill. Seedlings

Aiming at the production of robust chestnut plants suitable for the establishment of plantations or for their use in afforestation of forest ecosystems, we studied the effect of soil mixtures on growth, physiological characteristics, and nutritional status of chestnut seedlings. In a complete randomized-block design pot experiment, the soil used was derived from gneiss weathering (GN) and the treatments included GN and GN with fertilization (GN-FER) as well as mixtures of GN with forest floor (FF) from chestnuts (GN-CFF) and from evergreen broad-leaved trees (GN-EFF). Both types of forest floor and fertilization had similar positive effect on seedlings growth and photosynthetic parameters. The most significant effects on soil nutrients were found in GN-CFF and GN-EFF treatments, except for extractable phosphorus (P) and exchangeable potassium (K), whose highest concentrations were observed in GN-FER. Leaf analysis showed that the concentrations of almost all nutrients in all treatments were within the adequacy limits, with the exception of nitrogen (N) in GN, copper (Cu) in GN-EFF, and K in all treatments. Both types of forest floor satisfied the nutritional needs of chestnut seedlings to a large extent and their incorporation in this type of soil combined with limited addition mainly of K and to a lesser degree of P could enhance adequacy of these nutrients in soil and plants. Thus, forest floor application may be an appropriate alternative to integrated nutrient management for robust chestnut seedlings production with significant forestry


Introduction
European chestnut (Castanea sativa Mill.) and several other species of the genus Castanea have been grown in many countries and are exploited for timber and/or nut production (Arrobas et al. 2018). In the Mediterranean region, chestnut is found at a variety of altitudes, from sea level to over 1000 m. It prefers a mean annual rainfall of over 600 mm with no dry season, or a very short dry season of up to 3 months. Good quality of nut and high product yield require a mean monthly temperature above 10 °C for at least 6 months of the year (Fernández-López and Alia 2003). It is also an important species of the Balkan Peninsula with high ecological and economic value (Konstantinidis et al. 2008) and most chestnut forests are managed as coppice stands, mainly for timber and nut production. Except for the production of timber and nut on an annual basis, chestnut offers additional advantages such as preservation of the biodiversity of forest ecosystems and its adaptation to organic farming systems (Papaioannou 2018).
Although chestnut grows on a wide variety of soils, optimal conditions for this species are deep, moderately fertile, and acidic soils, pH 4.0-4.5 (Kerr and Evans 1993). Other authors consider that optimum pH is around 5.5 (Bourgeois et al. 2004). It does not thrive on limestone, preferring welldrained and nutritionally poor sites (Pereira-Lorenzo et al. 2012). In Greece, C. sativa exhibits preference for broad soil types, occurring in soils derived from a wide range of soft rocks such as schist, flysch, and gneiss that generally produce deep soils (Konstantinidis et al. 2008).
In chestnut plantations, research on tree nutrition as well as on fertility and management of soils on which they grow is generally limited (Freitas et al. 2021). Applying fertilizers to chestnut plantations is a relatively recent practice to improve soil fertility and enhance young tree growth. Studies conducted on chestnut seedlings grown in pots showed that they responded positively to the application of fertilizers, and therefore, similar results are expected in chestnut orchards (Laroche et al. 1997). Additionally, Pérez-Cruzado et al. (2011) evaluated the effects of the application of woodbark ash on the growth and the nutritional status of a 5-yearold hybrid chestnut plantation resulting in the improvement of the nutritional status of the plantation, mainly in terms of K, Ca, and Mg as well as in the increase of diameter and height of trees. Concerning chestnut forests, plant litter (forest floor) decomposition has long been recognized as an essential process for nutrient cycling and organic matter turnover, both being important determinants of plant productivity and ecosystem carbon storage (Hattenschwiler et al. 2005;Lensing and Wise 2007;Papaioannou et al. 2019).
Limited data on soil requirements and nutrition of chestnut in Greece makes necessary the research on the soil conditions and the amounts of nutrients required for its satisfactory growth under integrated management practices. In this context, the aim of the present research was to study the effect of soil mixtures, based on soil derived from gneiss weathering and two forest floor types, on growth, physiological characteristics, and nutritional status of chestnut seedlings. It is hypothesized that these forest floors will promote soil enrichment with nutrients and therefore will be effective in increasing chestnut seedlings nutritional status to produce robust plants, suitable for the establishment of plantations or for their use in afforestation of forest ecosystems. Introducing new and effective integrated nutrient management strategies that can promote nutrient availability to young chestnut seedlings will question the established practices of intensive agriculture based on nutrients replenishment by conventional inorganic fertilization which results in wellknown negative environmental impacts. So far, there are no data in the literature focused on the effects of organic fertilization on chestnut seedlings, whereas the nutrient status of large chestnut trees has been more extensively studied (Rodrigues et al. 2020a,b). Therefore, the results of our study are expected to be important in establishing a chestnut plantation based on integrated nutrient management practices reducing at the same time the environmental risk through the application of chemical fertilizers.

Chestnut Seedlings and Experimental Conditions
Nuts were collected for the purposes of the present study from Mount Pelion in Thessaly, Central Greece (indicative coordinates of the area are 39° 23′ 54″ N, 23° 6′ 53″ E).
After their collection, all the nuts were washed in order to select the most robust, while the seemingly unsuitable ones were removed. For the germination of the nuts, they were placed into peat and refrigerated at 0 °C until planting in the pots. The fruit moisture content, as a percentage of the fresh weight during the sowing date, after drying and repeated weighing until they reach a constant weight, was found to be 45%. The weight of each chestnut fruit was approximately 10 g (1 kg contained 102 nuts). The experimental part of the study, which included germination, planting, and the first growth stages of chestnut seedlings, was carried out in a greenhouse for the first 3 months. Afterwards, during the warmer months, the plants were transported to the farm of the Department of Forestry and Natural Environment of Aristotle University of Thessaloniki, under shady conditions (approximately 28% of sunlight was blocked) and remained there until the end of the experiment. The total duration of the experiment was eight months.
Planting took place in five (5) L pots. Materials used were: (a) Sandy loam mineral forest soil derived from gneiss weathering, pH 5.9 (b) Chestnut tree forest floor (c) Evergreen broad-leaved tree forest floor Both forest floor types were in natural litter form. Based on the above-mentioned type of soil and the two types of forest floor, the following treatments were applied in the pots: (i) Soil (control, GN) (ii) 2/3 soil + 1/3 evergreen broad-leaved tree forest floor (GN-EFF) (iii) 2/3 soil + 1/3 chestnut tree forest floor (GN-CFF) (iv) Soil + fertilizer Osmocote Topdress 19 (9% ΝΟ 3 -Ν and 10% ΝΗ 4 -Ν)-6-11 + 2MgO + 0.5Fe (ICL Specialty Fertilizers Americas) (GN-FER) The trials were arranged in a complete randomized-block design consisted of the four above-mentioned treatments replicated seven times and the pots were rotated to change position at least once a week. During the first 3 months of the experiment, when the seedlings were in the greenhouse, tap water was used for plant irrigation three (3) times per week and after the transportation of the pots in the farm, irrigation was accomplished with drilling water every morning for ten (10) min. Fertilization was carried out three (3) times during the experimental period using 10 g of fertilizer per pot each time.
Chemical properties of the soil derived from gneiss and the two types of forest floor, as well as the chemical analysis of the irrigation water, are presented in Tables 1, 2, and 3, respectively.

Plant Growth and Photosynthesis Measurements
At the end of the experiment, plant growth characteristics and plant physiological parameters were evaluated. In particular, plant growth characteristics included plant height, diameter, aboveground dry weight, root dry weight, and total dry weight. Moreover, photosynthetic quantum yield, chlorophyll content index (CCI), and net photosynthesis (CO 2 assimilation) were measured as follows. Photosynthetic quantum yield measurements were performed using the fluorometer Fluorpen FP100 (Photon System Instruments, Czech Republic). The leaves on which the measurements were made were healthy, not injured, at the same growth stage, and under similar orientation and exposure to sunlight. Two measurements were taken on leaves of different positions on each plant and in all repetitions of each treatment. Observations of CCI, as well as of net photosynthesis (CO 2 assimilation), were obtained under the same conditions. To measure CCI, the chlorophyll meter CCM-200 (Opti-Science) was used, which records the absorption of chlorophyll at two different wavelengths. CO 2 assimilation was performed using the portable instrument LCi Portable Photosynthesis System (ADC, BioScientific Ltd., England).

Leaf Analysis
After the removal from the pots, each seedling was separated into leaves, stem, and root and the fresh weight of each sample was measured. Leaves were washed with deionized water and were dried in the oven at 70 °C for 48 h and the dry weight was measured. After drying, all the leaves were ground in a laboratory mill to pass 40-mesh sieve and stored in small plastic containers. Total nitrogen (TN) was determined by the Kjeldahl method (Stevenson 1982). Powdered leaf samples were digested with H 2 SO 4 , HNO 3 , and HClO 4 (Allen et al. 1986). The solutions obtained from this procedure were analyzed for P spectrophotometrically by the molybdenum blue method, and for K, Ca, Mg, Na, Cu, Fe, Zn, and Mn by atomic absorption spectroscopy. Boron was determined with the azomethine-H method (Petridis et al. 2013) after dry combustion of leaf samples for 4 h at 500 °C and ash dilution with 0.1 N HCl.

Statistical Analysis
Statistical analysis was performed using the SPSS v.27 software. Univariate General Linear Model was performed to compare seedlings growth, soil parameters, and leaf nutrient content as well as the physiological parameters among the four experimental treatments. In this case, we were interested in testing the statistical hypothesis of the equality of a mean population set regarding the treatments as factor. Significant differences between the means of measured parameters of the treatments were tested at p ≤ 0.05 using Tukey's HSD test. Figure 1a and b present the means and standard errors of height (cm) and diameter (mm), respectively, while Fig. 2a-c present the means and standard errors of aboveground dry weight (g), root dry weight (g), and total dry weight (g) values, respectively, of all chestnut seedlings in the four treatments, measured at the end of the experiment. The height and diameter values of chestnut seedlings were significantly lower in GN treatment compared to GN-EFF, GN-CFF, and GN-FER ( Fig. 1a, b). The aboveground dry weight was significantly higher in GN-FER treatment compared to GN and GN-EFF, with GN treatment displaying the lower values (Fig. 2a). Root dry weight showed significantly lower values in GN treatment compared to GN-EFF, GN-CFF, and GN-FER, while an upward trend was observed in GN-CFF treatment among the three (Fig. 2b). Total dry weight measurements were found significantly higher in GN-EFF, GN-CFF, and GN-FER treatments compared to GN (Fig. 2c).

Soil Chemical Properties in the Four Experimental Treatments
pH values, TOC content, and nutrient concentrations in soil samples collected from all treatments are shown in

Photosynthetic Parameters of the Chestnut Seedlings in the Four Experimental Treatments
Measurements (means and standard errors) of photosynthetic quantum yield, chlorophyll content index (CCI), and net photosynthesis (CO 2 assimilation) are presented in Fig. 3a-c, respectively. No treatment affected the conversion of solar to chemical energy, as all values of photosynthetic quantum yield were high and there were insignificant differences among the four experimental treatments (Fig. 3a). In the case of CCI, significant difference was only recorded between GN and GN-FER treatments, with GN-FER treatment displaying the higher values (Fig. 3b). Regarding CO 2 assimilation of chestnut seedlings, as in the case of photosynthetic quantum yield, all the treatments showed high values and there were insignificant differences among them (Fig. 3c).

Discussion
Based on the enormous economic and ecological importance of chestnut and the demand of planting high-density chestnut orchards, introducing integrated nutrient management practices based on organic fertilization is critical to ensure higher yields. There was a significant increase in height, diameter, and biomass production of the chestnut seedlings in GN-CFF, GN-EFF, and GN-FER treatments compared to GN, and this is mainly attributed to the availability and adequacy of nutrients provided to soil by the decay of forest floor residues as well as to the nutrient release from the fertilizer Rodrigues et al. 2020b). Forest floor from both chestnuts and evergreen broad-leaved trees had positive effect on seedling growth, similar to fertilization. The results showed that both types of forest floor played a significant role in the nutrient availability for the chestnut seedlings. However, the values of all the measured plant growth parameters were higher in GN-CFF treatment compared to GN-EFF. Soylu and Serdar (2000), in an effort to produce drought resistant chestnut seedlings from 23 selected genotypes, in an area near the Black Sea, recorded at the end of the first vegetation period a maximum seedling height of 70.7 cm and a maximum diameter of 7.83 mm. The corresponding maximum values in the present study were 105.86 cm for the seedling height and 12.69 mm for the diameter. Improved growth of young chestnut plantation has also been reported According to the results of the present study, soil pH ranges from slightly acidic to slightly alkaline values (6.42-7.84) although it is well known that Castanea sativa is a calcifuge species and prefers acidic soils (pH 5.5) (Papaioannou et al. 2019). As gneiss weathering tends to form acidic and poorly developed soils, the initial pH value of the mineral forest soil of the current study was 5.9. The increased soil pH values in all treatments at the end of the experiment can be attributed to the pH of the irrigation water, which was 7.8. The application of the different forest floor types did not significantly influence the neutral nature of the soil mixture contrary to the fertilizer addition that affected pH, since the highest pH value was recorded in GN treatment, while the lowest in GN-FER. Previous research with nursery crops in Ontario has demonstrated the reduction of soil pH after the application of Topdressed Controlled-release Fertilizer (Osmocote Plus) as in the current study (Clark and Zheng 2017).
In general, positive impact of both GN-CFF and GN-EFF treatments in relation to GN and GN-FER was found concerning soil C, N, Ca, Mg, Na, Cu, Fe, Zn, Mn, and B concentrations, as well as C/N ratio. Incorporation of the two types of forest floor into the soil increased the addition of organic substrates and at the same time stimulated microbial activity, which enhanced organic matter mineralization and the release of available forms of the above-mentioned nutrients (Rodrigues et al. 2020a). Moreover, the low values of C/N ratio in GN and GN-FER treatments can be attributed to the corresponding lower TOC concentrations. On the other hand, only in the case of extractable P and exchangeable K, fertilization contributed significantly to the higher concentration levels of these macronutrients.
It is well known that the organic matter decomposition and release of ammonium nitrogen into the soil solution is a slow process compared to fertilization, which supplies readily available N for plant uptake. As the reported adequate level of soil TN for chestnut is > 0.1% (Freitas et al. 2021), TN concentrations in the treatments of our study are considered satisfactory for chestnut seedlings growth except for GN. Several studies have already showed the importance of N fertilization in chestnut tree nutrition (Arrobas et al. 2018;Rodrigues et al. 2020b). Raimundo et al. (2008) observed that soil TN concentrations in an 8-year-old chestnut plantation under biological production were 0.25% in the 0-5 cm depth, highlighting the significant role of chestnut litterfall on N dynamics.
The soil C/N ratio indicates the rate of organic residue decomposition and the availability of N. This ratio is relatively high in the forest floor and decreases as decomposition and humification proceed. A ratio of 20-30 results in an equilibrium state because N mineralization and immobilization occur at equal rates. The more favorable are the conditions for the rapid decomposition of plant residues, the lower C/N ratio and higher amount of available N are (Weil and Brady 2017). Fan et al. (2015) referred that the C/N ratio not only reflects soil fertility, but also can regulate plant growth and nutrition, while Zechmeister-Boltenstern et al. (2015) report that the C/N ratio depends on the nature and origin of the biomass. Although the underlying mechanisms controlling the combined roles of forest floor, soil properties, efficiency of nutrient cycling and microbiome on N content and C/N ratios remain unclear, Rodeghiero et al. (2018) using Boosted Regression Tree (BRT) models concluded that forest type is the most significant variable. In our study, the C/N ratio of both forest floor types belonging to broadleaves forest category was less than 30, which may have been the cause of its positive effect on soil N mineralization and the adequate supply of N to chestnut seedlings (Papaioannou et al. 2022;Zhang et al. 2022).
The concentrations of soil extractable P and exchangeable K were significantly higher in GN-FER treatment compared to the others. In the case of soil P, its concentrations in all the other treatments were rather low (< 6.0 mg kg −1 ), while Arrobas et al. (2017) in a study of soil conditions of a 4-year-old chestnut seedling plantation reported a much higher P concentration (15.48 mg kg −1 ) compared to GN, GN-CFF, and GN-EFF treatments of our study. The soil exchangeable K concentrations followed the same trend with that of P concentrations. Light-textured soils (< 35% clay) as with the soil of our study are generally poor in K-bearing minerals and therefore release low K by weathering, presenting at the same time low retention capacity (Sardans et al. 2008). According to Zimmermann et al. (2002), in mature chestnut clusters growing on gneiss soil in the region of Copera (Switzerland), the exchangeable K concentrations were found to be 0.63 cmol c kg −1 in the 0-6 cm of topsoil, while in the 20-85 cm depth, K concentrations were reduced even to 0.01 cmol c kg −1 . Ιn a report concerning the effect of chestnut forest management on the chemical properties of Mount Athos (Greece) gneiss soils as in our study, Papaioannou et al. (2011) reported soil exchangeable K concentrations between 0.53 and 0.64 cmol c kg −1 in the 0-10 cm depth. More recently, Arrobas et al. (2018) working with chestnut orchards in the district of Bragança, NE Portugal, pointed out that the high levels of soil exchangeable K were not linearly correlated to leaf K concentration. A mechanism of limited K uptake by chestnuts was proposed due to the low available soil moisture levels during the fruit growth phase under Mediterranean conditions. Exchangeable Ca concentrations in the soil of the present study are within the limits of adequacy for forest vegetation growth. According to Papamichos (2006), Ca deficiency rarely occurs in Greek forest ecosystems. In GN-CFF and GN-ΕFF treatments, soil Ca concentrations were approximately twice higher compared to GN, revealing the critical role of forest floor on Ca availability. Our data are partially in agreement with those of Papaioannou et al. (2011), who found soil exchangeable Ca concentrations between 3.67 and 12.32 cmol c kg −1 in chestnut clusters growing on gneiss soils of Mount Athos. On the contrary, except for GN and GN-FER treatments, significant lower Ca values were observed in the Castanea orchards and forest soil profiles (0-80 cm) located in Griva on Mount Paiko (northern Greece) due to the strongly acidic nature of soil derived from mica schist (Papaioannou et al. 2022). Exchangeable Mg concentrations in the soil were also adequate for the chestnut plant growth in all treatments, with the highest value determined in GN-EFF, followed by GN-CFF treatment. Exchangeable Na concentrations in the soil ranged from 0.30 to 0.54 cmol c kg −1 . These concentrations, combined with the high pH values of the substrates, were probably due to the water quality. Portela and Louzada (2012) for a 15-year-old chestnut orchard in a shallow soil in Portugal found Na concentrations of 0.15 cmol c kg −1 in the soil profile, a value significantly lower than those of the present study.
Concerning soil micronutrients, the highest concentrations of Fe, Zn, and B were measured in both GN-CFF and GN-EFF treatments, while those of Cu and Mn were exclusively noted in GN-CFF and their differences from the other treatments were significant. In the cases of Fe, Zn, Cu, and Mn levels, in GN-CFF and GN-EFF treatments, all concentrations were within or above the adequacy limits. Data from other research works have shown much lower concentrations of available Fe (between 3.74 and 5.72 mg kg −1 ), and higher available Cu concentrations (1.49-2.85 mg kg −1 ) in chestnut plantations of Turkey in the 0-30 cm soil depth (Toprak and Seferoğlu 2013); significant higher available Mn concentrations (39-116 mg kg −1 ) in soil profiles derived from mica schist of chestnut orchards and forests of Griva on Mount Paiko, Greece (Papaioannou et al. 2022), and Zn concentrations of 1.57-3.27 mg kg −1 in the 0-20 cm depth of soils of the Black Sea area (Turkey) derived from igneous or sedimentary rocks (Dengiz et al. 2011). These differences support the critical role of soil parent material, as well as of soil pH on the availability and uptake of micronutrients by chestnuts as recently reported by Rodrigues et al. (2020a,b). Incorporation of both types of forest floor in the soil had positive impact on B increase, although there are no relative references about adequacy limits for this nutrient. Portela and Louzada (2012) reported soil B concentrations of 0.44 mg kg −1 in the topsoil (0-20 cm) and 0.10 mg kg −1 in the 20-40 cm depth in a chestnut plantation in Portugal, while Rodrigues et al. (2020b) observed values up to 3.31 mg kg −1 in the surface (0-10 cm) of a loamy-sand soil after the application of Exactyon AG 18:5:13, a controlledrelease fertilizer.
Leaf TN was found within the limits of adequacy in all cases, except for the GN treatment, and the higher concentrations were presented in GN-FER treatment. It is nevertheless noteworthy that leaf TN concentrations in the treatments with the two forest floor types did not stand out so much from those of the fertilizer treatment. Santa Regina (2000) investigated the alteration of leaf N concentrations during the vegetation period in a young chestnut forest growing in granite soil and found values between 8.2 and 28.5 g kg −1 . P concentrations are within the adequacy limits in all treatments and the higher value was noticed in GN. Toprak and Seferoğlu (2013), while were investigating the seasonal alterations in leaf P content in thirty chestnut orchards in Turkey, measured comparable concentrations with the current study ranging between 1.4 and 1.9 g kg −1 .
In contrast to N and P, leaf K concentrations are below the adequacy limits. Similar results were presented by Laroche et al. (1997) who measured low K values (5 g kg −1 ) in the leaves of chestnut seedlings without being fertilized, while these values did not change significantly after fertilization. On the other hand, increased K values (6.5-15.6 g kg −1 ) in chestnut leaves were reported by Santa Regina et al. (2001) in young chestnuts, 7-30 years old, in regions of Italy, France, and Spain. Moreover, higher K concentrations (7.1-9.4 g kg −1 ) compared to the results of the present research were measured by Sariyildiz and Anderson (2005) in chestnut leaves in southwest England and by Papaioannou et al. (2022) in chestnut orchards and forest sites in northern Greece.
Both leaf Ca and Mg concentrations in all treatments were within the adequacy limits. Significantly reduced Ca concentrations (7-9.6 g kg −1 ) in chestnut leaves during September were reported by Sariyildiz and Anderson (2005) for clusters growing in fertile soils derived from different parent materials in southwest England. Additionally, according to Portela et al. (2010), who studied the Mg adequacy in chestnut leaves in Northern Portugal, the lowest content in robust trees was 1.8 g kg −1 , while the highest value in trees with symptoms of deficiency was 1.5 g kg −1 .
The current study, in which Ca and Mg concentrations in the leaves were within the sufficiency range in the treatments of both forest floor types, supports the hypothesis that the forest floor has a positive effect on the nutritional status of the chestnut seedlings. Leaf Na concentrations ranged from 0.28 to 0.43 g kg −1 . Higher Na concentrations (3.5-6 times) in leaves were presented by Papaioannou et al. (2022), who studied the leaf nutrient content in chestnut forests and orchards, in northern Greece.
Concerning all the micronutrients (Fe, Mn, Zn, Cu, and B), their concentrations in the leaves were in general within the adequacy limits in all the experimental treatments. It is possible that the higher leaf Fe values observed in GN treatment were due to the intense uptake by chestnut seedlings and the accumulation of Fe in the leaves since plant growth in this treatment was limited compared to the others. Slightly higher Fe concentrations (197-271 mg kg −1 ) in leaves were measured by Toprak and Seferoğlu (2013) in thirty chestnut orchards in Turkey. Lower leaf Mn concentrations (450-739 mg kg −1 ) were recorded by Papaioannou et al. (2017) in 10-, 20-, and 40-year-old chestnuts growing on gneiss soils in Mount Athos, Greece. Significantly higher leaf Cu concentrations (6-28 mg kg −1 ) were referred by Anagnostakis et al. (2012) in research conducted in USA on 1-to 12-year-old chestnut trees, as well as by Toprak and Seferoğlu (2013) (16-24 mg kg −1 ) in chestnut orchard in Turkey. In another research by Arrobas et al. (2017), they reported Zn concentrations between 25 and 44 mg kg −1 in leaves of young chestnut trees in NE Portugal. Several studies have demonstrated the importance of B for chestnut in the context of a significant increase in nut productivity and the adaptation of trees in climate changes (Arrobas et al. 2018;Freitas et al. 2021). In this study, the leaf B concentrations were in the adequate range in all treatments while in a 10-year study concerning boron adequacy in chestnut leaves in Portugal, it was concluded that B was frequently inadequate in concentrations below 20 mg kg −1 (Portela et al. 2007).
No treatment affected photosynthetic quantum yield, as there were insignificant differences among them. Moreover, there are no published data on the quantum yield of chestnut leaves. In most plants, values of quantum yield under 0.75 mean that the plant is under stress, while stress is not observed at values close to 0.8 (Maxwell and Johnson 2000). Our data show that quantum yield values were below 0.75 and this could be explained by the fact that the measurements were taken in September, where the high temperatures combined with the increased evapotranspiration caused stress on the seedlings.
In the case of CCI, significant difference was only recorded between GN and GN-FER treatments reflecting the nutritional status of the chestnut seedlings. The highest CCI value measured in GN-FER treatment was probably due to the highest N content in the seedling leaves because of fertilization (El Kohen and Mousseau 1994; Arrobas et al. 2018). Finally, concerning CO 2 assimilation, all the values were high and similar, indicating high photosynthetic capacity of chestnut plants in all the experimental treatments.

Conclusions
This study provides novel information on the effects of different integrated nutrient management strategies on the growth and nutritional status of chestnut seedlings. In particular, the present paper pointed out that the incorporation of forest floor from chestnut trees as well as from evergreen broad-leaved trees in a poor nutrient soil derived from gneiss weathering had significantly more remarkable effects on the growth of chestnut seedlings compared to initial gneiss soil, but similar effects with the fertilized soil. Both types of forest floor satisfied the needs of chestnut seedlings for almost all nutrients and could be an alternative to fertilization practice to produce robust seedlings, suitable for the establishment of plantations or for their use in afforestation of forest ecosystems. According to this study, incorporation of forest floor in this type of soil could be combined with limited addition mainly of potassium and to a lesser degree of phosphorus in order to enhance adequacy of these nutrients in soil and plants.
In the context of an integrated nutrient management for the production of robust chestnut seedlings, the use of forest floor as a part of a mixture with a gneiss-derived soil could be considered a beneficial management practice with significant forestry and environmental impacts. Future research on different type of soil:forest floor ratios (i.e., ½ soil + ½ forest floor etc.) and/or on forest floors of different origin could also indicate other equally or more appropriate soil mixtures for chestnut plantation. Moreover, the results suggest that longterm studies should be performed to better clarify the benefits of using forest floors in chestnuts production.
Funding Open access funding provided by HEAL-Link Greece.

Conflict of Interest The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.