Introduction

Hyrcanian forests are unique forests that have maintained their remnants from the last ice age (Sagheb-Talebi et al. 2014). The Caspian vegetation region on the shores of the Caspian Sea, form a green belt, 110 km wide and 800 km long, of temperate deciduous trees. This ecoregion receives 600 to 2000 mm of rainfall per year. Beech forests are part of the ancient, valuable forest ecosystems in the northern hemisphere because the beech trees have been created by natural regeneration and belong to the third geological period. Based on published statistics, this species alone comprises 23.63% of the number of forests in northern Iran and 29.96% of the volume (Sefidi et al. 2016; Azaryan et al. 2021). In mixed forests, the composition of a mature stand is determined by various dynamic processes that influence the establishing and developing of a stand (Levula et al. 2003). The overstorey composition of trees also significantly affects litter quality and topsoil properties (Kemner et al. 2021; Kim et al. 2021; Wang et al. 2021; Qin and Wang 2022).

Soil directly contributes to various ecosystem functions and services including net primary production, climate regulation, nutrient cycle and carbon sequestration (Singh et al. 2018; Parhizkar et al. 2021). Soil is a finite resource because it develops over very long periods of time. On the human timescale, soil can be considered a non-renewable natural resource. Plant community composition at the ground level alters soil processes and functions through a variety of factors including microclimate change, bedding and root secretions, and habitat or resource provisions for soil microbial communities (Lin et al. 2021). Therefore, understanding the impact of plant communities on multiple soil functions is of great value.

The fertility of forest soils is seen in its ability to provide nutrients for forest trees (Nguemezi et al. 2020). The forest canopy cover can affect soil fertility as an important part of forest sites. Amount and quality of litter (Majasalmi and Rautiainen 2020), through-fall and stem flow, rooting patterns, microbial activities and forest floor processes are typically affected by soil fertility, directly or indirectly (de Vries et al. 2021), and result in different levels of forest productivity (Majasalmi and Rautiainen 2020).

Soil quality is promoted through ongoing organic material input and decomposability via microbial activities that generate soil fertility hotspots. These processes lead to an increase in soil microbial diversity and respiration under the tree canopy (Wang et al. 2022) and create a microhabitat for increasing biodiversity (Catenazzi and Donnelly 2007; Rathore et al. 2022). These hotspots of fertility have been hypothesized to form through both biotic and abiotic processes (i.e., litterfall and decomposition) (Charley and West 1977), atmospheric deposition (Fenn et al. 2003) and microbial activity (Žifčáková et al. 2016; Wang et al. 2022) that make feedback loops and fortify high nutrient accumulation (Schlesinger et al. 1996). The attributes of the fertility hotspots will also differ based on factors such leaf quality, species composition and the ecosystem (Alameda et al. 2012). For instance, the canopy cover of trees provides a suitable environment for microbial colonization (Ortiz et al. 2022) and activity by reducing solar radiation, temperature and soil water evaporation (Berry et al. 2013). Changes in forest tree composition due to global climate change also obviously affect ecosystem performance (Mueller et al. 2012). Hence, the impact of trees on biogeochemical cycles has widely been studied, but the results have been inconsistent (Wang et al. 2018, 2021; De Andres 2019).

Beech species can accelerate soil acidification and thus leaching of nutrients, decreasing topsoil fertility. Therefore, the presence of other tree species in beech forest stands may help improve soil characteristics and nutrient cycling (Kooch 2012). In the forests of northern Iran, the contributions of the other tree species that grow with the old-growth beech trees to microbial activity, nutrient cycles, and belowground properties have not been considered yet. Here, we thus assessed the effect of mixed forest stands in composition with different tree species on soil functional indicators in organic and mineral layers. We hypothesized that the soil fertility and microbial hotspots would predominantly correlate with the tree species and litter properties. Our findings on soil functions will serve as a base to optimize forest structure and improve ecosystem services.

Materials and methods

Study area

The Golband region (study site), consisting of 36,855 ha in northern Iran (Fig. 1A), is characterized by uneven-aged stands (i.e., tree diameters between 10–150 cm). The Golband watershed lies between 51°17′ E to 51°46′ E, 36°27' N to 36°35 N. The mean altitude is 2000 m above sea level, mean total rainfall is 900 mm, and mean annual temperature is 11 °C. The slope of the region varies between 5 and 70%. According to the American USDA Soil Taxonomy classification (Hughes et al. 2017), the soil of the region is Alfisols. The area consists of mixed beech forest dominated by oriental beech (Fagus orientalis Lipsky), wingnut (Pterocarya fraxinifolia Lam.), oak (Quercus castaneifolia C. A. Mey.), lime tree (Tilia begonifolia Stev.), maple tree (Acer velutinum Boiss.), Caucasian zelkova (Zelkova carpinifolia Dippe), ash (Fraxinus excelsior L.), hornbeam (Carpinus betulus L.), Cappadocian maple (Acer cappadocicum Gled), and Caucasian alder (Alnus subcordata C.A. Mey.). Less-frequent species (< 10%) are elm (Ulmus glabra Huds.), wild cherry (Prunus avium L.), and wild service tree (Sorbus torminalis Crantz). Herbaceous species such as Asperula odorata L., Hypericum androsaemum L., Euphorbia amygdaloides L., and Polystichum sp. cover more than 85% of the forest floors (Anonymous 2018).

Fig. 1
figure 1

Location of the study area (Golband Forest) in the Mazandaran Province, northern Iran, with 100 studied patches (plots) of dominated tree species (A, B). Soil samples (0–10 cm depth) were taken in a 25 × 25 cm area under each canopy cover (C). Note: Fagus: Fagus orientalis Lipsky, Quercus: Quercus castaneifolia C. A. Mey., Pterocarya: Pterocarya fraxinifolia Lam., Tilia: Tilia begonifolia Stev., Zelkova: Zelkova carpinifolia Dippe, Acer C.: Acer cappadocicum Gled, Acer V.: Acer velutinum Boiss., Fraxinus: Fraxinus excelsior L., Carpinus: Carpinus betulus L., and Alnus: Alnus subcordata C.A. Mey

Full size image

Sampling design and laboratory measurements

Patches (hereafter plots) of dominant tree species were identified in the study area. In total, 100 plots (10 replications for each tree species) were considered (see Fig. 1A, B). Each plot includes an individual tree (DBH about 50 cm) of the dominant tree species that are always surrounded by similar tree species. All plots were located between 1000 and 1100 m a.s.l., had a similar aspect (north), slope class (12%–16%), forest management (preserved areas and intact without harvesting) and were at least 1000 m from each other. In September, a litterbag experiment was set up to assess litter decomposition in the field (Wieder and Lang 1982). Four litter traps (1 m × 1 m) were placed in each plot at the four sides of the studied tree species at a distance of one-third of the crown radius from the stem, with 40 bags of laboratory-dried litter from each species within each plot. In August, soil samples (25 cm × 25 cm × 10 cm; see Fig. 1C) were taken at the site of each litter trap. In total, 400 (i.e., 10 tree species × 10 replicated plots × 4 samples) litter and soil samples were gathered, the litter and soil samples within each plot (n = 4) were bulked separately to yield one composite litter and one composite soil sample for each plot (total: 100 litter and 100 soil samples) for analysis. In addition, litter (forest floor or O-horizon) thickness was measured at the litter trap sites (Dechoum et al. 2015).

An elemental analyzer (Fisons EA1108, Milan, Italy) was used to measure total C in samples and nutrient components in litter (Parsapour et al. 2018). Bulk and particle densities (BD and PD, respectively) of soils were determined with clod (Plaster 1985) and pycnometer (Blake and Hartge 1986) methods. Soil porosity was computed as 1 − (BD / PD) (Pires et al. 2014). The distribution of aggregate size (0.053 and 0.25 mm for microaggregates and 0.25 and 0.50 mm for macroaggregates) was determined as described by Cambardella and Elliott (1992). Aggregate stability and texture of soils were determined using the Yoder and Bouyoucos methods (Bouyoucos 1962; Kemper and Rosenau 1986). Soil pH was determined using an Orion Ionalyzer Model 901 pH meter in a 1:2.5, soil: water solution. EC (Electrical Conductivity) was determined using an Orion Ionalyzer Model 901 EC meter in a 1:2.5 soil: water solution. Soil organic C and total N were determined using the Walkley–Black and Kjeldahl methods (see Allison 1975; Bremner and Mulvaney 1982). C and N sequestrations in this study were computed as C or N sequestration = C or N content × Soil depth × BD × 0.1, where 0.1 is a conversion factor. Particulate organic C and particulate organic N (POC and PON) were measured using physical sundering (Cambardella and Elliot 1992). Dissolved organic C and dissolved organic N (DOC and DON) was analyzed using the procedure of Jones and Willett (2006). Soil extraction solutions were used for the colorimetric determination of NH4+ (at 645 nm) and NO3− (at 420 nm) concentrations (Li et al. 2014). Soil available K, Ca, and Mg were measured using atomic absorption spectrophotometry (Bower et al. 1952) and available P using spectrophotometry (Homer and Pratt 1961). The separated fine roots (i.e., diameter < 2 mm) were dried at 70 °C, then weighed (Neatrour et al. 2005).

Soil water content and temperature (as soil climate variables) and biota were measured in summer (15 August) and fall (15 November). Soil water content was measured by drying soil samples at 105 °C for 24 h. Soil temperature was measured using a digital probe-thermometer sensor (TFA Dostmann, Model 30.1048, Ottersberg, Germany) in the field (Zancan et al. 2006). Before drying earthworms were picked from the soil samples and categorized into ecological classes based on exterior specifications (Kooch et al. 2014). Extraction and Acari and Collembola counts were obtained using the Berless-Tulgreen funnel method; to obtain mesofauna counts, a certain amount of soil was weighed and placed in the Berless-Tulgreen funnel; after 4 days, mesofauna were collected in 0.5% v/v formalin and counted) (Page 1986). Nematodes were extracted using tray (Whitehead and Hemming 1965) and centrifugation (Niknam 1991) methods, then fixed and transferred to glycerin. Densities of soil protozoa were determined using an extraction method and light microscopy (Mayzlish and Steinberger 2004). Bacterial and fungal counts on nutrient agar and potato dextrose agar were obtained by serial dilution and plate count methods (Kooch et al. 2020). Soil basal respiration (BR) was measured using CO2 emission method, and CO2 absorption was measured in an alkaline solution (Alef 1995). Substrate-induced respiration (SIR) (based on CO2 production; Anderson and Domsch 1990) and microbial biomass C, N and P (i.e., MBC, MBN and MBP) were measured using a fumigation-extraction method (Brookes et al. 1985). The metabolic quotient (qCO2), microbial ratio or entropy and C availability index were calculated based on the stoichiometry of soil organic C, respiration and microbial biomass C (Insam and Domsch 1988; Anderson and Domsch 1990; Cheng et al. 1996). Soil enzyme activities were calculated using the Schinner and von Mersi (1990) protocol. Soil C and net N mineralization (Cmin and Nmin) were measured as described by Raiesi (2012a, b a, b) controlled laboratory conditions.

Statistical analyses

Normality of data was assessed using a Kolmogorov–Smirnov test and homogeneity of variance was checked using Levene’s test. One-way analysis of variance (ANOVA) was used to compare litter and soil properties among different tree species; means that differed significantly were analyzed using Duncan’s test (P ≤ 0.05). All analyses were done using SPSS (version 20; IBM, Armonk, NY, USA). In addition, principal component analysis (PCA using PC-Ord version 5.0; Mc Cune and Mefford 1999) was performed to identify any patterns in the changes in litter and soil characteristics between different tree species.

Results

Litter properties

Litter chemistry differed significantly among the various tree species. The contents of NPK, Mg and Ca in the A. subcordata litter was highest. Litter thickness was greatest under F. orientalis (≈ Q. castaneifolia). Litter C was significantly higher under F. orientalis, Q. castaneifolia and P. fraxinifolia, which had the same functional traits. In addition, F. orientalis had the highest litter C/N ratio (Table 1). The highest litter decomposition rates after 360 days were found under Alnus, but the trend in litter decomposition was the same for all tree species over the experiment. The ANOVA results revealed that significant differences in litter decomposition were due to litter types of the various tree species. At all sites, the litter lost half of its initial mass during the incubation period (Fig. 2 and Table S1).

Table 1 Mean (± SE; n = 10) of litter properties under different tree species
Full size table
Fig. 2
figure 2

Litter mass remaining (kg ha−1) under different tree species. Data in details are presented in Appendix 1. Note: Fagus: Fagus orientalis Lipsky, Quercus: Quercus castaneifolia C. A. Mey., Pterocarya: Pterocarya fraxinifolia Lam., Tilia: Tilia begonifolia Stev., Zelkova: Zelkova carpinifolia Dippe, Acer C.: Acer cappadocicum Gled, Acer V.: Acer velutinum Boiss., Fraxinus: Fraxinus excelsior L., Carpinus: Carpinus betulus L., and Alnus: Alnus subcordata C.A. Mey

Full size image

Soil properties

All soil properties, except soil density, the amount of sand and C sequestration, were affected by the tree species. The most available P and K was found under Alnus, whereas litter under A. subcordata ≈ C. betulus species had the most available Ca and Mg. Soil organic C, POC and DOC were significantly higher at sites having F. orientalis and Q. castaneifolia than at the other sites, but soil C/N ratio was highest under F. orientalis, Q. castaneifolia and P. fraxinifolia. Fine root biomass was highest in the topsoil of A. subcordata ≈ C. betulus and the lowest under F. orientalis (Table 2). In fact, these two species had similar values for some soil characteristics. Overall, the various tree species affected the soil microclimate and biotic characteristics (Table 3). In addition, soil biota population differed significantly among the tree species, and seasonal changes in all traits evaluated were similar among the tree species. The soil water content was higher under F. orientalis ≈ Q. castaneifolia, with maximum values during the autumn. In the summer, soil temperatures were highest in A. subcordata ≈ C. betulus ≈ F. excelsior stands (Table 3), which have the same function as other species. Except for soil bacteria and fungi, which were more abundant in the summer, the activities of other soil biotas were higher in the autumn when an ideal soil climate (i.e., higher soil water content and lower soil temperature) prevailed for most of the studied soil organisms (Table 3, Fig. 3).

Table 2 Mean (± SE; n = 10) of soil properties under different tree species
Full size table
Table 3 Mean (± SE; n = 10) of soil climate and biota in summer (S) and fall (F) seasons under different tree species
Full size table
Fig. 3
figure 3

Variability of soil biota related to soil climate (summer and fall) under different tree species

Full size image

The different tree species exerted significant effects due to the habitat types on the activity of ecological groups of earthworms. Maximum activity was observed under A. subcordata trees. The densities of soil acarina were highest under A. subcordata ≈ C. betulus ≈ F. excelsior, whereas greater densities of collembola, nematode and bacteria were observed under A. subcordata species. Moreover, higher densities of protozoa were detected under A. subcordata and C. betulus and more fungi under A. subcordata ≈ C. betulus ≈ F. excelsior ≈ A. velutinum. (Table 3). The effects of trees on changes in soil microbial and enzymatic activities were significant (Table 4). Soil BR, SIR, MBN, MBP, qCO2, urease and acid phosphatase activities were highest under A. subcordata. Arylsulfatase activity was highest under A. subcordata ≈ C. betulus and invertase activity highest under A. subcordata ≈ C. betulus ≈ F. excelsior. The soil under F. orientalis had a higher MBC, while the highest MBC/MBN ratio was measured for F. orientalis or Q. castaneifolia stands (Table 4). After 17 weeks of incubation, the soil C mineralization in the litter was differentially affected by the tree species in rank order of A. subcordata ≈ C. betulus ≈ F. excelsior > A. velutinum. > A. cappadocicum. > Z. carpinifolia ≈ T. begonifolia > P. fraxinifolia ≈ Q. castaneifolia ≈ F. orientalis. However, soil N mineralization after 35 days, was differentially affected by tree species in rank order of A. subcordata > C. betulus > F. excelsior ≈ A. velutinum. > A. cappadocicum. ≈ Z. carpinifolia > T. begonifolia > P. fraxinifolia ≈ Q. castaneifolia ≈ F. orientalis (Fig. 4; Table S2).

Table 4 Mean (± SE; n = 10) of soil microbial and enzyme activities under different tree species
Full size table
Fig. 4
figure 4

Mean (± SE; n = 10) of soil C and N mineralisation (A, B) under different tree species. Data in details are presented in Appendix 2. Note: Fagus: Fagus orientalis Lipsky, Quercus: Quercus castaneifolia C. A. Mey., Pterocarya: Pterocarya fraxinifolia Lam., Tilia: Tilia begonifolia Stev., Zelkova: Zelkova carpinifolia Dippe, Acer C.: Acer cappadocicum Gled, Acer V.: Acer velutinum Boiss., Fraxinus: Fraxinus excelsior L., Carpinus: Carpinus betulus L., and Alnus: Alnus subcordata C.A. Mey

Full size image
Fig. 5
figure 5

PCA based on the correlation matrix of the tree species (A), litter and soil properties (B, C). Note: Fagus: Fagus orientalis Lipsky, Quercus: Quercus castaneifolia C. A. Mey., Pterocarya: Pterocarya fraxinifolia Lam., Tilia: Tilia begonifolia Stev., Zelkova: Zelkova carpinifolia Dippe, Acer C.: Acer cappadocicum Gled, Acer V.: Acer velutinum Boiss., Fraxinus: Fraxinus excelsior L., Carpinus: Carpinus betulus L., and Alnus: Alnus subcordata C.A. Mey

Full size image

Relationship among trees with litter and soil properties

In the PCA analysis of the 70 variables evaluated for litter and soil samples, two principal components (PC1 and PC2) explained over 55% (PC1 = 50.13%, PC2 = 6.90%) of the total variance. The PCA outcomes showed a clear discrimination in the litter and soil properties among tree species due to functional traits or habitat types of trees. Two categories of drivers for soil fertility and microbial activities were revealed Fig. 5. Group 1 (F. orientalis, Q. castaneifolia, P. fraxinifolia, T. begonifolia, Z. carpinifolia, A. cappadocicum. and A. velutinum.) enhanced soil microbial biomass of carbon and had a positive effect on soil properties, whereas group 2 (F. excelsior, C. betulus and A. subcordata) showed litter nutrients and enhanced biota activities, C and N cycles (Fig. 3A − C).

Discussion

Litter properties

Our findings clearly indicate litter chemistry and decomposition were differentially affected by the various tree species. In addition to the quality of litter, the availability and content of nutrients that are returned to the soil environment are fundamental for soil fertility and optimum tree growth (Cao et al. 2020). Houle et al. (2015) believed that the type of tree species and the quality of their litter determine the amount of available nutrients and the mechanism of litter decomposition. Chen et al. (2020) pointed to the importance and role of forest species in restoring nutrients to the soil and stated that the litter quality (thickness and litter elements like C, N, K, P, Ca, Mg and C/N ratio) strongly affects decomposition rate and soil fertility. In this study, we showed that A. subcordata provides more P, K, Ca, Mg, and total N in the litter than the other species. In addition to the nutrients, the C/N ratio is an important indicator in decomposition dynamics (Goh and Totua 2004; Vesterdal et al. 2012; Kooch and Bayranvand 2019). The litter of A. subcordata species had a higher N content than that of the other species. As a consequence, the C/N ratio was lower. We therefore hypothesize that A. subcordata releases more C into the soil than the other species do. A. subcordata is a pioneer species in Hyrcanian forests that adds new organic substances to the forest soil layers every year (Koupar et al. 2011). It also forms a symbiosis with N-fixing actinomycetes and fixes atmospheric N into the soils, increasing soil fertility and providing good quality litter (i.e., lower C/N ratio and higher values of N, P, K, Ca and Mg) (Taleshi et al. 2009; Parsapour et al. 2018). The high N content of A. subcordata litter also increases the populations of soil organisms and the mineralization rate of elements (Glaser et al. 2018). The higher C/N ratio in F. orientalis litter, compared to the other species, may cause a higher recalcitrance and lower decomposition rate than for litter of the other tree types.

Soil properties

As mentioned earlier, soil properties differ significantly among tree species (Wang et al. 2021). The trees drive biogeochemical regulation in ecosystems via the stabilisation of organic C among other things. The lower amount of organic C under A. subcordata and C. betulus is the result of a rapid mineralisation (Kooch 2012; Błońska et al. 2018), which is related to the occurrence of fertility hotspots. In comparison to contents in other trees, the higher contents of N of A. subcordata and C. betulus improve the soil N (Sayyad 2009). The content of soil macro elements is also associated to the release of nutrients by trees and nutrient cycling in the forest floor (Dijkstra and Smits 2002; Osborne et al. 2020). Humus formation and nutrients cycling can also be affected by the canopy of trees (Majasalmi and Rautiainen 2020). A. subcordata and C. betulus can lead to an increase in soil pH and fertility (Zeng et al. 2014; Majasalmi and Rautiainen 2020), whereas Q. castaneifolia and F. orientalis species have greater acidifying capabilities (Augusto et al. 2002) than other deciduous trees do.

Soil biological activities were generally lower under F. orientalis. At A. subcordata stands, the higher quality of the forest floor enhanced soil biota populations (Knops et al. 2002; Osborne et al. 2020). Our data indicated a temporal change in earthworms, acarina, nematodes, protozoa, collembola and densities of fungal and bacterial populations due to seasonal environmental changes (Chaudhuri and Paliwal 2008; Suthar 2012). Soil moisture as an important factor affects the physiology of microorganisms directly, affecting access to water and regulating access to organic matter, which in turn affects macroorganismal and microbial soil populations (Andrade et al. 2017). Soil temperature and water content are major drivers of variations in soil biota under different trees (Lozano-Parra et al. 2015). Higher soil temperature and less topsoil water under various tree species are unsuitable for epigeic activity in the summer. The faster reaction of epigeic earthworms than the other ecological groups is due to the higher sensitivity of epigeic earthworms to the soil microclimate compared to anecic and endogeic forms (Lagerlof et al. 2002), which anecic and endogeic groups can move to various soil layers when conditions are favourable (Nuutinen and Butt 2009). Higher soil fauna activities in the fall have also been confirmed (Crumsey et al. 2013; Ren et al. 2018). In the fall season, maximum earthworm activity was recorded among tree species having favourable soil water. The lower earthworm population in the summer is a result of the drier and warmer conditions (see Fig. 2). Crumsey et al. (2013) reported significant effects of soil water on the comparative frequency of earthworm species comparative frequency of earthworm species and that earthworm species richness was mainly regulated by soil water content rather than pH or soil organic C. Similar results were reported by Suthar (2012); low physiological activity and high temperatures in summer limit the activity of soil organisms.

Differences in soil fertility can also be the result of differences in the earthworm, acarina, collembola, nematode and protozoa populations and preferences (Sackett et al. 2013; Sigurdsson and Gudleifsson 2014). Rich stands with more nutrients and litter having a low C/N ratio (Rehschuh et al. 2021) are preferred by earthworms. Several studies (Drouin et al. 2016; Tucker Serniak 2017; Zhang et al. 2020) revealed a crucial influence of the chemical composition of plants (such as N, lignin, and phenols) on soil properties and fauna. According to our results, the lowest earthworm density and biomass was found at F. orientalis plots having a higher C/N ratio. Desirable conditions for earthworms were provided by A. subcordata, which had the lowest C/N ratio. Thus, soil bacteria activity was higher in A. subcordata plots and more fungi were present in plots with A. subcordata, C. betulus, F. excelsior and A. velutinum. Variations in physical properties (e.g., lower aggregate stability) and soil fertility can also alter bacterial and fungal populations in F. orientalis and Q. castaneifolia plots (Kim et al. 2021). Previous studies showed provisional schemas of soil fungi and bacteria population in various trees and indicated that soil bacteria were more abundant during the summer than the fall (Kuffner et al. 2012; Preusser et al. 2019; Li et al. 2021). Generally, the soil organismal activities in A. subcordata plots might be attributable to higher litter quality and more favourable topsoil temperatures (Glaser et al. 2018). Bacteria are more sensitive to a low pH. Therefore, the highest biomass can normally be related to neutral to slightly alkaline conditions of the soil at A. subcordata stands (pH > 7). Hence, there is a positive correlation between bacterial biomass and pH. Differences in quantity and quality of the litter influences microbial and enzymatic activities, nutrient accessibility and, thus, all biogeochemical cycles (Zheng et al. 2018). Compared to other species, the highest value of BR, SIR, MBN, MBP, and qCO2 was recorded in soils at A. subcordata stands. Tardy et al. (2014) showed a negative effect of low soil fertility on BR and SIR, similar to our findings. The results of Sasongko et al. (2019) revealed that an increase in forest soil nutrients can lead to higher microbial activities, BR, and SIR in the soil.

Our results showed significant differences in enzymatic activities among the tree species. Soil enzymatic activities are affected by soil management strategies and trees (Silva et al. 2012) and can be used to predict changes in soil quality (Guo and Han 2008; Yao et al. 2020). Higher enzymatic activity causes faster decomposition and higher availability of organic nutrients (Wang et al. 2013). The clay fraction seems to contribute to the accumulation of soil enzymes via their stabilisation and protection (Zhong et al. 2015) under A. subcordata. In soils at the A. subcordata sites, the activity of urease increases with increasing pH, EC, total N, and nutrients (Cheng et al. 2013) and decreases with higher C/N ratios. Acid phosphatase activities also can significantly differ depending on the tree species (Chodak et al. 2021) and are strongly affected by pH, soil water, total N and organic C content (Wang et al. 2021). The activity of sulfatase is correlated with soil particle size and highest in the clay fraction (Ling et al. 2014) under A. subcordata. Greater soil water can reduce the revival oxidation potential anaerobic conditions of the soil that constrain enzyme activities (Brockett et al. 2012) under F. orientalis. Low sulfatase activities at the F. orientalis stands can also be a result of low soil pH (Wang et al. 2016). Invertase plays a key role in the N and C cycles by hydrolyzing sucrose into glucose and fructose (Zhong et al. 2015). A. subcordata gives rise to a higher decomposition rate of litter and accelerates N cycling and invertase activity compared to F. orientalis (Zhong et al. 2015). In addition, higher pH (Li et al. 2009) and better fertility (Zhong et al. 2015) at A. subcordata stands improves the invertase enzyme activity. According to our findings, the soil at A. subcordata, C. betulus and F. excelsior stands had the highest mineralisation rate of C and N in comparison with other species. Parallel with our data, previous researches (Eickenscheidt et al. 2014; Uri et al. 2014; Tarighat and Kooch 2018) pointed out that mineralization of soil C and N by N-fixing tree species is significantly higher than for non-N-fixing species. In general, some habitat characteristics such as the quality of litter and soil physicochemical characteristics influence the variability of soil C and N mineralisation under various tree species. In general, A. subcordata provides better conditions for the mineralization of soil C and N due to the more alkaline soil.

Relationship among tree litter types and soil properties

This study is the first scrutiny that quantifies the impacts of typical trees in the Hyrcanian mixed beech forest on litter and soil properties. The PCA revealed a clear difference in the specific litter and soil properties among the studied trees. The type of broadleaf species also affects spatial variations in nutrient cycling. The better quality of organic matter under A. subcordata plots increased the activity of soil fauna/flora, increasing soil fertility. Our comparisons indicate that soil quality increases in order of F. orientalis < Q. castaneifolia < P. fraxinifolia < T. begonifolia < Z. carpinifolia < A. cappadocicum. < A. velutinum. < F. excelsior < C. betulus < A. subcordata. Also, C. betulus and F. excelsior serve an essential role in modifying microbial flora and soil nutrient content. Thus, a mixed natural forest is fundamental to providing soil services in temperate ecosystems and is a pivotal for sustainable forest care. Since the tree species influences litter quantity and quality, forest habitats provide various functions. Therefore, forestry management should be in the direction of ecosystem tolerance and select the most suitable tree species to ensure proper forest care.

Conclusions

The tree species in the Hyrcanian mixed beech forests—F. orientalis, Q. castaneifolia, P. fraxinifolia, T. begonifolia, Z. carpinifolia, A. cappadocicum, A. velutinum, F. excelsior, C. betulus, and A. subcordata—strongly influence litter and soil properties. Our findings revealed that the differences in basic characteristics of these 10 species resulted in a distinct effect on microbial flora and affected soil nutrient cycling. Soil fertility and microbial hotspots in these forests were clearly species-specific confirming our research hypothesis of soil fertility and microbial hotspots governed by tree species and litter properties. According to our data, A. subcordata (as N-fixing species), C. betulus and F. excelsior species are the main drivers of microbial activities related to nutrient cycling in old-growth beech forests. In fact, the admixture of valuable broad-leaved species with beech stands at fertile sites served as a significant silvicultural system for maintaining soil quality via natural or human-induced soil acidification. These findings improve our knowledge of the impacts of different tree species on the litter and soil properties of deciduous mixed forests and subsequent productivity of the relevant ecosystem.