The Soil Environment of Abandoned Charcoal Kiln Platforms in a Low-Altitude Central European Forest

Abstract

This study examines the soil environment of eight charcoal kiln platforms and the neighboring soil in Czech and Bohemian low-altitude forest stands. Both mixed and undisturbed soil samples were used to assess the hydrophysical soil properties, nutrient content, cation exchange capacity, enzyme activity, and soil active carbon content, while soil color, stoniness, root density, and horizon thickness were estimated in the field. Charcoal-rich horizons had high total organic carbon concentrations and total nitrogen content (about 150% and 40% higher than in the organomineral horizons of the control plot, respectively), with total carbon stocks being higher than those in neighboring forest soils. Fine root density was highest in the charcoal-rich horizons, encouraged by high soil porosity, aeration, and favorable chemical properties. Enzyme group activity differed between individual soil horizons and kiln and control plots, with depolymerization enzyme activity highest in charcoal-rich horizons and humification enzyme activity highest in mineral horizons. Phosphatase, chitinase, and phenoloxidase activity were highest in charcoal-rich horizons, while β-glucosidase activity remained similar across horizons. After long-term abandonment, kiln sites shift from inhospitable sites to localized hotspots for plant and microbial growth, having more favorable physical, enzymatic, and chemical soil properties than the surrounding areas. This study confirmed that kiln production platforms act as microhabitat hotspots, also providing information on a wide spectrum of soil properties linked with soil microorganisms and root growth.

1. Introduction

Various anthropogenic factors can affect forest soil sites, including the deposition of charcoal following the establishment of charcoal kilns (hereafter also referred to as hearths) within the forest. The existence of such features is inextricably linked to the history of European forests, with kilns having been created and restored to produce charcoal, using wood harvested from coppices, for centuries. Nowadays, the remains of such hearths can be found in their thousands in many areas, often with a density of more than one kiln site per hectare [1].

Kiln sites today are mostly found within forested landscapes, the forest’s long-term generation turnover ensuring that the sites remain relatively undisturbed. In contrast, while it is likely that kiln platform sites existed in present-day agricultural land, their stratigraphy has most likely been disturbed or destroyed through arable land management. In the forest, autonomous topsoil development overlays and conserves the original kiln platform layers with the new organic horizons typical of most forest communities. As such, kiln soils can be co-defined by (1) soil bodies that create a set of pedogenic and overlying humus horizons, and (2) forest communities, the existence of which is conditioned by the surrounding soil environment [2]. Such sites can be further defined edaphically, based on the soil’s sorption capacity and its bound nutrient content. Finally, all these factors will be further affected by local climatic conditions, the site´s trophic state, and the local water regime [3].

The soil’s origin and its quality are expressed through defined soil properties, with a specific nomenclature that is used to identify the formation of individual soil horizon sequences. In the case of charcoal kilns, the soil is classified taxonomically under Spolic Technosols [4], which differ from the surrounding soil environment. The kiln’s impact on the soil can be defined in two ways: first, through the establishment of the charcoal kiln production platform, and any subsequent logging or continuous biomass removal within the short-term generation cycle of the forest [5]; and second, through the soil environment’s developmental trajectory away from the kiln production platform, which initially represents an inhospitable environment for most plants. Generally speaking, such soils will be subject to alterations in nutrient balance, environmental alkalization, an increase in carbon concentration and its transition into a stable form of pyrolytically decomposed organic matter (black carbon), and soil nitrogen depletion, most of which will be volatilized at relatively low temperatures [6]. As such, kiln platforms represent hotspots of different soil properties, with the differentiated accumulation or depletion of individual elements [7]. Consequently, numerous studies have focused on the importance of such hearths as plant habitats, for example, their effect on forest stand production [8] and the stability of forest ecosystems, including the provision of ecosystem services [8,9,10].

The establishment and operation of a kiln platform, despite being highly localized and often a one-off activity, can cause severe changes to the local soil chemistry, resulting initially in an overall reduction in biological activity and substantial nutrient disbalance. Thus, for some time, the soil around the kiln platform will differ from that in the surrounding area, being characterized as an inanimate substrate with limiting conditions for most organisms [11,12]. Over the course of time, however, the character of the Technosols will be restored and a new hydric, chemical, physical, and biological balance will be established, such that the platform will be capable of hosting plants and other organisms once again. Paradoxically, the restored environment may actually become even more hospitable than the surrounding soil, mainly due to a change in carbon forms and an increase in the area’s concentration of black carbon [13].

The newly restored soil balance stands out in the forest environment due to its more autonomous topsoil development. This surface forest soil is characterized by a unique and distinctive arrangement of horizons, the genesis of which leads to an increase in horizon thickness and count [14]. In the mineral soil horizons, pedogenesis via partial processes causes the release of elements from crystalline bonds. This is particularly noticeable in relation to the various forms of aluminum that are largely bound to humic substances, the concentrations of which will be affected by factors such as local vegetation type and site conditions [15,16]. However, kilns create local disturbances that cause alterations in forest soil stratigraphy, particularly in the topsoil, which plays an essential role through its ability to store and gradually release water [17,18], optimize water–air conditions, and balance the variously bound nutrients in the organic soil components [19].

In addition to its role in the water–air relationship and exchangeable nutrient storage, soil is also a fundamental reservoir of organic matter in different forms. “Non-living” soil organic carbon [20] represents approximately two-thirds of all carbon in the terrestrial environment [21], stored in various forms [22]. This includes the so-called black carbon [23], which is found in enhanced concentrations at charcoal kiln platform sites and is characterized by a high rate of sorption for water and nutrients and the ability to optimize the extreme physical properties of the soil, e.g., by adjusting the soil’s retention abilities and air regime.

Biologically fixed organic carbon is stored in the living soil biota, which produces various enzymes that are crucial for maintaining nutrient cycles and nutrient availability, and for promoting organic matter decomposition and soil metabolic processes [24,25]. Unlike the chemical and physical properties of the kiln soil environment, the biochemical nature of kiln platform soils is still largely unexplored. Nevertheless, there is some evidence that both the activity and diversity of soil microbiota can increase over time following the abandonment of such kilns, particularly in acidic edatopes [26].

In recognizing the relative scarcity of studies examining the pedochemical features of kiln platforms in forest habitats, particularly as regards their biochemical properties, the aim of this treatise was to evaluate and compare the soil environment of charcoal kiln platforms with that of the surrounding soil, in terms of site condition and enzyme activity. Based on a study of the available literature, we hypothesize that such kilns will provide favorable soil conditions for plant roots and soil microorganisms, due to the associated layer of pyrolytically decomposed biomass.

2. Materials and Methods

2.1. Study Sites

This study was conducted at eight localities in the Czech Republic (

Table 1. Locations of the charcoal kilns examined in this study, including data on altitude, size, slope, and forest type for each forest site.

Kiln Site	Study Area	GPS Coordinates	Altitude (m a.s.l.)	Size - Fall Line (m)	Size-Contour Line (m)	Slope (°)	Current Forest Type
1	Brno	N 49°14′0.67″ E 16°49′33.13″	446	4	6	8.8	mixed
2	Brno	N 49°13′34.81″ E 16°49′3.73″	390	6.5	8.5	19.6	mixed
3	Brno	N 49°20′44.54″ E 16°39′44.17″	346	10	12.5	9.3	mixed
4	Brno	N 49°20′22.15″ E 16°40′58.34″	446	10	12	9	mixed
5	Brno	N 49°25′11.13″ E 16°46′24.38″	549	10.5	13	8.6	mixed
6	Brno	N 49°25′24.82″ E 16°46′3.29″	586	12	12	7.7	mixed
7	Prague	N 49°55′8.609″ E 13°45′15.592″	475	10	11	16	mixed
8	Prague	N 49°54′55.7″ E 13°44′52.2″	489	11	10	15	mixed

), situated in two lowland forest areas with a high concentration of charcoal kilns, the first being situated approx. 20–30 km northeast of Brno (Brno area) and the other approx. 35–45 km southwest of Prague (Prague area) (Figure 1).

The conditions at each site were comparable, each having a similar slope inclination (Brno 7–20°, Prague ca. 15°), altitude range (Brno 346–586 m a.s.l., Prague 470–500 m a.s.l), kiln platform diameter, and common tree species composition. Natural conditions at both study areas were typical of lower altitudes, with an average annual temperature range of 8.2–8.7 °C, 550–700 mm average annual precipitation, and a long vegetation season of around 150–165 days. Geologically, the Brno area was characterized by a predominance of greywacke or granodiorite rock with a loess loam admixture, with loam predominating on less hilly terrain. In contrast, the Prague area was typified by a Barrandian dark grey silty shale bedrock alternating irregularly with spilites, with a slight loess loam admixture predominating.

Based on a preliminary kiln inventory, six study localities were chosen in the Brno area and two in the Prague area (Table 1, Figure 1). After verifying the presence of a charcoal layer, any kiln close to a forest road or that had been damaged by machinery, wood transport, or tree uprooting was rejected. In the case of the kiln platforms, soils were classified as Eutric Cambisols (Protoabruptic, Siltic, and eventually Protostagnic in the case of thicker loess loam layers), with higher stoniness and no hydromorphological features in the Prague study area. In terms of the neighboring soils, brunification processes prevailed, along with various anthropogenic influences, such as pedoturbation, loam accumulation, and charcoal admixture. Forest tree species composition was typically mixed, comprising deciduous species of lower altitudes and warmer (drier) climates, such as oak (Quercus spp.), European hornbeam (Carpinus betulus L.) and, to a lesser extent, European beech (Fagus sylvatica L.), with a coniferous admixture of Norway spruce (Picea abies/L./H. Karst.), European larch (Larix decidua Mill.) and Scots pine (Pinus sylvestris L.) of anthropogenic origin.

2.2. Field Survey

Edaphic and geological conditions at each site were assessed using soil probes inserted into the kiln production platform and control sites in the neighboring soil along a contour line situated 10–15 m from the kiln. Each probe was marked and determined in accordance with the current World Reference Base [4]. The soil horizons thereby determined were then aggregated according to their type or mode of formation as: (i) organic (O) horizons of humification, with an organic carbon content > 20%; (ii) organomineral (OM) A horizons underlying organic layers, with an organic carbon content < 12.5%; (iii) organomineral carbonic (OMC) material, with the accumulation of black carbon (typically >10% of organic carbon and the presence of charcoal); (iv) mineral red (MR), with mineral horizons underlying OMC and being affected by high temperatures during ignition, leading to a red coloration; (v) mineral (M), with mineral horizons in intact profiles or in buried soil layers under the kiln. Horizon thickness was assessed as the difference between the lower and upper limits, measured from the soil surface in cm. Soil color was defined according to the Munsell color system. Fine root density in each horizon was measured in roots per 100 cm³.

The soil was excavated to a depth of 40 cm, following the individual horizons, and two types of samples were taken, i.e., undisturbed (core) samples (100 cm³ volume) for the determination of its hydrophysical properties (three repetitions per horizon), and mixed samples (ca 500 g weight) for the determination of its chemical and biochemical properties. The samples for the biochemical properties were transported to the laboratory at 4 °C and then frozen at −28 °C until further analysis. In total, 24 undisturbed samples and eight mixed samples were collected from each kiln horizon and control plot, providing a total of 120 undisturbed samples and 40 mixed samples for the kilns and 72 undisturbed samples and 24 mixed samples for the control sites, the differences in the number of samples taken being the result of different topsoil organization in the kiln and control plots (Figure 2).

2.3. Laboratory Analysis

Undisturbed samples were analyzed to assess the bulk density (ρ_d) (g cm⁻³) of oven-dried (105 °C) samples; maximum capillary capacity (Θ_MCC) and retention water capacity (Θ_RWC) (2- and 24-h water suction from a fully saturated sample, respectively); permanent wilting point [27] [% vol.] using the equation Θ_PWP = Vhk, where Vh is hygroscopic moisture (calculated as the air-dry water content, multiplied by 1.8) and k is the empirical constant given by the soil texture (2 for clay, 2.5 for sandy clay loam, clay loam, and silty clay loam, and 3.0 for sand, loamy sand, sandy loam, loam, silt loam, and silt textural classes); utilizable water capacity, using the equation UWC = Θ_RWC − Θ_PWP [mm] for 200 mm of soil horizon column; minimum aeration capacity, using the equation A_MCC = P – Θ_MCC [% vol.], where P is porosity expressed as P = (ρ_s − ρ_d)/ρ_s 100 [% vol.], with ρ_s representing specific density [g cm⁻³], assessed using a Gay–Lussac pycnometer [28].

After passing through a 2-millimeter sieve, the mixed samples were assessed for active and potential soil reactions, using pH/H₂O with distilled water and pH/KCl with a 0.2 M KCl solution, respectively, with a sample:eluent ratio of 1:2.5 (w:v); hydrogen cation (H⁺) concentration (mmol₊ kg⁻¹), obtained using dual pH measurements [29]; mobile aluminum content (Al³⁺) (mmol₊ kg⁻¹), according to Sokolov [15]; oxalate aluminum content (Al_o) (g kg⁻¹) [30]; total aluminum content (Al_t) (g kg⁻¹), assessed after mineralization in hydrofluoric acid, followed by FAAS (mg kg⁻¹) [30]; available mineral nutrients (Ca, Mg, and K) (mg kg⁻¹) from the BaCl₂ leachate [31]; cation exchange capacity (CEC) (mmol₊ kg⁻¹), obtained using the summation method; base saturation (BS) (%), derived using the equation BS = (Ca²⁺ + Mg²⁺ + K⁺)/CEC); acid saturation (AS) (%), derived using the equation AS = (H⁺ + Al³⁺)/CEC); and phosphorous content (P) (mg kg⁻¹), obtained using the spectrophotometry of a solution of ascorbic acid, H₂SO₄, and Sb³⁺. Using a pulverized sample passed through a 0.25-millimeter sieve, we also assessed total organic carbon (TOC) (%) content, acquired spectrophotometrically in chromosulfuric acid [32], and total nitrogen (TN) (%) content, measured using the Kjeldahl method [33].

The selected soil properties were expressed through nutrient content; nutrient ratios (e.g., Al_o/Al_t and C/N) and sums (e.g., the sum of UWC, TOC, TN, and Ca, Mg, K, and Na contents, using horizon bulk densities and thicknesses to a depth of 40 cm).

Microtitration fluorometric tests using black 96-well microplates were used to determine the extracellular enzyme activity. All hydrolytic enzyme activities were determined using a 1 g sample of soil suspended in 50 mL of distilled water, which was then sonified for four minutes to disturb the soil particles. Next, 200 μL of the soil suspension was added to 50 μL of methylumbelliferyl, a substrate solution for β-glucosidase, sulphatase (arylsulphatase), phosphatase and chitinase (N-acetyl-glucosaminidase) detection. In addition, 50 μL of 7-aminomethyl-4-coumarin substrate solution was used for the determination of leucine-aminopeptidase [34]. Further samples with concentrations of 50, 100, and 300 μM of each fluorogenic substrate were also tested and those with the highest activity were selected for further analysis. The small plates were incubated for 120 min at 20 °C. Fluorescence quantification took place using an Infinite M NANO⁺ microplate reader (TECAN, Lyon, Germany), using an excitation wavelength of 360 nm and an emission wavelength of 465 nm.

Peroxidase and phenoloxidase (both oxidative enzymes) activity was gauged photometrically using L-DOPA (l-3,4-dihydroxyphenylalanine) as a substrate [34] and 400 μL of the filtered soil suspension, mixed with 100 μL of acetate buffer and 100 μL of the L-DOPA solution, 20 μL of 0.3% H₂O₂ (v/v) having been added to the L-DOPA solution to determine the peroxidase activity. After 18 h of incubation in the dark at 20 °C, the absorbance was measured at a wavelength of 450 nm. The sums of β-glucosidase, sulphatase (arylsulphatase), phosphatase, and chitinase (N-acetyl-glucosaminidase) enzyme activity were expressed in terms of the activity of depolymerization enzymes, while the sums of the peroxidase and phenoloxidase activity were expressed as the activity of humification enzymes.

Soil-active carbon (permanganate oxidizable carbon (POxC)) content in the soil samples was assessed according to the method used by the authors of [35], whereby 2.5 g of air-dried soil was mixed with 18 mL of distilled water and 2 mL of 0.2 M KMnO₄. After 2 min of shaking and 10 min of sedimentation, 0.5 mL of the supernatant was mixed with 49.5 mL of distilled water and the suspension was transferred to a 96-well microplate. Absorbance was measured at a wavelength of 550 nm.

2.4. Data Processing

Using RStudio v. 2022.07.1, the data were processed to compare the parameters, clustered using two variable groups, i.e., soil horizon type (O, OM, OMC, MR, and M) and habitat type (kiln (K) and control (C)). After testing normality with the Shapiro–Wilk test, nutrient stocks in individual habitats were compared for K vs. C, using the parametric Student’s t-test or a non-parametric Mann–Whitney test, based on a significance level of alpha = 0.05. Principal components analysis (PCA) was carried out using the ‘vegan’ package, v. 2.5-6 [36], after data standardization, using the ‘scale’ function. The dataset was reduced by applying PCA using the following continuous variables: root density, A_MCC, ρ_d, TOC, P and K content, BS, Alo/Alt, and the sum content of the depolymerization and humification enzymes. Relationships between the continuous variables and horizon types in ordination diagrams were expressed using the ‘ordispider’ function. If the p-value of the one-way ANOVA was <0.05 using the Tukey HSD test, we applied a multiple comparisons procedure, with graphical expression performed using the ‘ggplot2′ package, v. 3.3.6 [37]. Error bars in the multiple comparison plots were expressed as standard deviation.

3. Results

3.1. Soil Stratification

There were clear differences in soil stratigraphy between the kiln deposition layers and the formation of soil-genetic horizons in neighboring forest soils. In the forest soil, for example, the horizons were arranged in a sequence that is typical for cambisols (i.e., O, OM, M metamorphic, and substrate horizons), with a mostly dark greyish-brown to brown color gradient (organomineral-10YR 4/2-3; metamorphic-7.5YR 4/4; substrate-10YR 5/6). In the kiln, however, the soils were classified as Spolic Technosols (Carbonic, Pyric, and Thaptotransportic). They contained a dark charred OMC horizon, ranging in color from black (10YR 2/1) to a greyish-brown (10YR 4/2) below the newly formed OM horizon, with a typically red-hued (2.5YR 4/6) MR layer deposited below this. Finally, the lowest layer consisted of a sequence of original M soil layers of a camel-brown color (10YR 5/6) previously affected by the mechanical treatment of the production platform. In both the kiln and control plots, the O and OM horizons were of a similar thickness (2–3.5 cm and 4–5.5 cm, respectively). OMC horizons were, on average, 10.5 cm thick, and those of the MR horizons were 9 cm thick.

While rooting density was similar in the O, OM, and M horizons in both the kiln and control plots, much higher rooting density levels were recorded in the OMC horizon, with levels even exceeding those in the humification horizons by up to 70% (Figure 3,

Table 2. Mean values (x¯ ) and standard deviation (sd) of soil parameters in treatment soil horizons.

			Kiln					Control
Parameter	Units		O	OM	OMC	MR	M	O	OM	M
Root density	pcs. dm−2	x¯	86.7	60.0	147.9	19.3	11.5	72.9	90.3	24.4
		sd	15.7	35.6	16.1	11.2	12.0	11.5	27.6	11.5
ρd	g cm−3	x¯	0.41	0.85	0.71	1.31	1.58	0.64	0.91	1.43
		sd	0.12	0.25	0.05	0.28	0.13	0.02	0.32	0.19
ΘRWC	% vol.	x¯	48.10	31.92	34.91	29.11	27.08	62.67	28.12	32.06
		sd	17.68	2.58	4.23	5.09	3.40	4.77	13.04	12.27
ΘPWP	% vol.	x¯	8.73	25.30	19.98	12.99	14.66	6.46	8.08	13.89
		sd	1.16	3.55	2.70	6.13	8.26	1.03	2.87	8.96
AMCC	% vol.	x¯	3.67	6.40	9.61	4.78	3.17	0.99	4.98	3.33
		sd	3.24	2.02	2.96	2.11	1.67	0.21	2.02	1.84
pH/KCl	-	x¯	4.03	3.67	3.59	3.89	3.74	3.70	3.44	3.75
		sd	0.54	0.64	0.27	0.62	0.24	0.77	0.45	0.35
CEC	mmol kg−1	x¯	249.74	112.65	159.48	72.81	72.79	191.46	101.73	76.60
		sd	87.27	38.49	65.35	23.86	26.02	16.29	26.94	31.27
BS	%	x¯	84.7	49.0	42.8	61.7	72.5	71.8	39.9	45.5
		sd	19.6	30.9	19.8	28.0	22.5	20.3	20.7	24.3
TOC	%	x¯	27.56	8.07	13.85	1.68	0.70	28.06	5.51	1.06
		sd	10.10	5.34	4.20	1.12	0.28	6.38	2.77	0.76
TN	%	x¯	1.18	0.32	0.47	0.11	0.08	1.40	0.34	0.09
		sd	0.35	0.17	0.11	0.04	0.02	0.15	0.17	0.03
C/N	-	x¯	23.00	23.33	30.50	15.00	9.40	20.40	16.00	11.33
		sd	3.27	5.03	8.04	5.24	3.58	4.62	2.39	4.92
Alo/Alt	-	x¯	0.078	0.048	0.100	0.017	0.011	0.063	0.030	0.023
		sd	0.036	0.031	0.013	0.009	0.003	0.018	0.010	0.009
P	mg kg−1	x¯	44.29	13.00	24.50	13.75	12.20	82.60	15.50	14.93
		sd	20.70	4.36	27.97	8.99	6.14	43.42	13.16	17.56
Mg	mg kg−1	x¯	258.9	112.0	91.6	123.6	208.2	246.8	120.6	143.7
		sd	90.07	11.53	46.71	75.62	120.44	21.26	57.53	141.22
Ca	mg kg−1	x¯	2674.1	648.3	978.0	570.9	684.0	1329.4	492.8	450.8
		sd	1440.8	159.3	776.9	351.2	427.8	565.2	403.1	513.2
K	mg kg−1	x¯	299.6	111.0	70.3	126.0	121.8	502.0	141.1	73.3
		sd	69.3	44.4	24.8	55.1	44.0	204.0	77.8	29.3

), with low standard deviations in OMC indicating a relatively uniform rooting density between sites, compared with the variable values observed in the control plots.

3.2. Soil Chemical Properties

While soil reaction values tended to be lower in the control O and OM horizons, there were no significant differences between the control and kiln sites (Table 2). In both the kiln and control sites, CEC was highest in the O horizon (followed by OMC), presumably due to its higher colloidal soil fraction, as was the level of carbon and nitrogen concentration. However, while the C/N ratio was highest in the O horizon at the control sites, the highest levels were observed in the OMC horizon (OMC > OM > O) at kiln sites, presumably due to the higher carbon concentration in this horizon. Likewise, while P, Mg, Ca, and K levels were all highest in the O horizon at all sites, the levels dropped in the OM horizon but increased again in the OMC, MR, and/or M horizons at the kiln sites. The kiln OMC horizon also had the highest levels of non-bound Al relative to total content.

3.3. Soil Biochemical Properties

Aside from peroxidase, phenoloxidase, and humification enzyme levels, which were all highest in OMC, MR and/or M horizons, all other biochemical properties were highest in the O horizon (

Table 3. Biochemical soil properties with means (x¯) and standard deviations (sd) of enzyme group activity and permanganate oxidizable carbon content (POxC), within horizons and treatments.

Parameter	Units		Kiln					Control
			O	OM	OMC	MR	M	O	OM	M
β-glucosidase	nmol g−1 hr−1	x¯	3507.0	452.0	350.6	86.7	59.2	4886.5	544.5	162.4
		sd	1999.1	373.3	161.7	81.4	46.8	2945.0	256.3	156.0
Chitinase	nmol g−1 hr−1	x¯	929.2	55.6	98.8	38.7	88.7	642.2	145.4	102.4
		sd	421.6	37.7	52.3	31.3	31.7	66.9	96.9	91.5
Leucine-aminopeptidase	nmol g−1 hr−1	x¯	105.3	20.6	32.2	11.4	4.3	70.3	57.5	19.3
		sd	90.5	13.4	30.5	11.0	3.0	93.1	67.2	30.0
Phosphatase	nmol g−1 hr−1	x¯	6607.7	476.0	841.1	164.3	95.5	4599.1	762.3	187.3
		sd	3021.9	354.0	413.9	119.1	39.8	1191.3	323.4	107.9
Arylsulphatase	nmol g−1 hr−1	x¯	147.9	27.2	38.9	10.2	10.4	98.9	49.1	23.9
		sd	52.6	35.4	34.8	8.3	9.7	60.7	39.9	27.2
Depolymerization enzymes	nmol g−1 hr−1	∑	11,297.0	1031.0	1362.0	311.0	258.0	10,297.0	1559.0	495.0
		sd	5248.0	759.0	562.0	209.0	128.0	2117.0	478.0	309.0
		x¯	2259.4	206.2	272.4	62.2	51.6	2059.4	311.8	99.0
Peroxidase	nmol g−1 hr−1	x¯	362.4	832.7	848.5	867.2	874.0	374.3	770.1	847.7
		sd	206.2	96.2	132.2	66.1	102.4	47.3	111.2	355.6
Phenoloxidase	nmol g−1 hr−1	x¯	142.2	80.5	174.6	83.1	117.8	94.8	94.1	100.0
		sd	81.5	8.7	56.4	34.0	71.4	75.4	36.9	80.4
Humification enzymes	nmol g−1 hr−1	∑	505.0	913.0	1023.0	950.0	992.0	469.0	864.0	948.0
		sd	226.0	91.0	156.0	50.0	144.0	44.0	90.0	394.0
		x¯	252.5	456.5	511.5	475.0	496.0	234.5	432.0	474.0
POxC	mg C kg−1 dry mass	x¯	2483.6	1353.1	1075.1	291.0	177.4	2841.4	1183.0	205.0
		sd	686.1	257.8	383.4	149.9	98.8	18.4	497.8	149.1

). Overall, aside from peroxidase and β-glucosidase, kiln O horizons displayed higher enzymatic activity than the control sites. By contrast, leucine-aminopeptidase, phosphatase, and glucosidase enzyme activity in the control OM and M horizons was several times higher than that in the equivalent horizons in kiln plots. Chitinase, leucine-aminopeptidase, and phosphatase activities were significantly higher in kiln O horizons, with the activity of all three enzymes combined being nearly 70% higher than in the control sites. At both the kiln and control sites, enzyme group activity differed with soil depth, with depolymerization enzyme activity decreasing with depth (with a temporary peak in the kiln OMC horizon) and humification enzymes increasing with depth. While the POxC levels were slightly higher in the O horizon at the control sites, levels were generally higher in the kiln OM and M layers.

3.4. Utilizable Water Capacity and Nutrient Sums

While mean values for the sum of UWC down to 40 cm were similar at the kiln and control sites, with both sites always having values of >40 mm, the kiln sites displayed more variability (Figure 4a). The sum of the TOC was markedly higher at kiln sites, in some cases more than twofold higher, as was the sum of TN, which is largely dictated by soil organic matter (Figure 4b,c). Mean values for the sum of Ca and K (Figure 4f,g) were significantly higher at kiln sites, with levels of other macrobioelements remaining similar between sites. Aside from Ca, nutrient sums tended to show less variability at kiln sites (i.e., a narrower quartile range), especially the summarized contents of P and Na (Figure 4d,h). While there were clear differences between kiln and control sites for the sum of UWC and sums of most nutrients, only the sum of C showed significantly higher levels at kiln sites (

Table 4. Sums of utilizable water capacity (sum–UWC) and nutrients at 0–40 cm depth. Statistically significant values (p < 0.05) are shown in bold italic.

Parameter	Units	p-Value S–W Test	Kiln		Control		p-Value t-Test
			Mean	Median	Mean	Median
sum-UWC	mm	0.646	54.3	53.7	57.2	57.1	0.6522
sum-TOC	kg m−2	0.664	17.5	19.0	10.0	11.1	0.0331
sum-TN	kg m−2	0.099	0.77	0.87	0.65	0.70	0.3409
sum-P	g m−2	0.003	53.1	46.1	90.0	49.5	0.9497
sum-Mg	g m−2	0.140	58.4	49.4	73.9	54.1	0.5404
sum-Ca	g m−2	0.033	325.6	296.9	236.4	122.8	0.2284
sum-K	g m−2	0.514	47.7	46.2	39.6	40.1	0.2284
sum-Na	g m−2	0.014	19.0	17.5	27.6	19.4	0.9497

).

3.5. Multivariate Interrelationships

Based on PCA analysis, there appeared to be a strong relationship between the chosen soil properties and certain soil horizons (Figure 5; see also the Supplementary Material). For example, while the O horizons of both the kiln and control plots were linked with a high accumulation of organic carbon and depolymerization enzyme activity, the mineral horizons were more closely linked with humification enzyme activity. The O horizons were also associated with higher available nutrient content and BS values. Unsurprisingly, both the M and MR horizons were linked with increasing bulk density values, while both the TOC and the Alo/Alt were more closely linked with the upper O horizons. This association of Alo with the upper organic layers was rather surprising, not least as the relatively low pH levels could result in Al toxicity, which could inhibit fine root growth [38]. However, it would appear that the increased Alo concentrations were buffered in exchangeable buffer zones [39] as a result of high CEC levels and sufficient base saturation (Table 2).

4. Discussion

4.1. Topsoil Formation, Stratigraphy, and Base Characteristics

Numerous studies have examined the importance of forest charcoal kiln platforms in relation to local history, nutrient supply, and different vegetation types. Such studies have typically shown that kilns were historically situated on hillside locations that a priori provided optimal soil trophy and hydric conditions for the growth and renewal of forest stands. Schmidt et al. [8] also confirmed a link between kiln production platforms and natural conditions in the local habitat, particularly as regards altitude, edaphic, and climatic conditions. At the peak of charcoal production, kiln sites were typically associated with the presence of naturally growing oak, beech, and hornbeam [11] in slightly acidic habitats with mesotrophic soils at lower altitudes (and lower vegetation zones). These same links were also confirmed in our own study.

The actual stratigraphy of kiln site soils will depend on the nature of disturbance associated with production platform construction and the subsequent wood charring. Hirsch et al. [40] reported that the bulk density of layers rich in pyrolyzed woody material was usually about 0.7 g cm⁻³, and this finding was confirmed in our study; however, in our study, the pH was lower (ca. 1.5). In this case, it is likely that pH was influenced by the different species compositions in the two studies, rather than from any amelioration effect on the soil, with the sites studied by Hirsch et al. [40] typically comprising maple (Acer pseudoplatanus), birch (Betula pendula), or aspen (Populus tremula), while our study sites were dominated by oak and hornbeam with an admixture of acidifying conifers, such as Norway spruce [41]. This is also likely to be the explanation for the higher TN levels (up to 10 times) in the charcoal-rich horizons at our sites. Other factors influencing these differences, other than the quality of pyrolytically processed organic matter, include 8 times higher C-levels in our study and the fact that a substantial proportion of N was fixed with organic matter [42]. Furthermore, in both cases, the length of time since the last production of charcoal remains unknown (potentially falling anywhere between 120 and 300 years previously [43]); thus, we do not know at what point along the trajectory of chemical balance reestablishment our investigation falls. This is further complicated by the effects of historical atmospheric deposition, with more recent years typified by an increase in atmospheric N compounds [44]. Finally, it is also likely that the use of different analytical methods in the two studies played a substantial role.

4.2. Rooting and Nutrient Content in the Charred Layers

The presence of charcoal in the soil had a positive effect on rooting in the OMC layer (Figure 3; Table 2), with levels significantly higher than those in the soil layers above or below. A similar beneficial effect on European beech seedling root growth was also demonstrated by Carrari et al. [11], though not in the case of seedlings of holm oak (Quercus ilex L.) or Turkey oak (Quercus cerris L.). An overall positive effect on plant growth has also been demonstrated for crops grown with biochar [45,46]. In these cases, the main cause of better plant growth was probably an increased nutrient and N availability [47]. Indeed, in our own study, we also confirmed the higher availability of some nutrients, especially C and P, as well as an increase in the C/N ratio. While the higher C/N ratio could potentially have resulted in reduced N availability [48], the existing high total N levels in the soil prevented any subsequent impacts on root growth.

Increased levels of elements such as Ca, Mg, and K can also have a negative effect on plants, potentially lowering the osmotic potential in the soil solution and inducing a physiological drought in plants [49,50,51]. In our study, permanent wilting-point values were highest in the charcoal-rich topsoil layers (Table 2), which is likely to have contributed to a lower total sum of utilizable water capacity values at the kiln sites (Figure 4a). As a result, it is also necessary to assess the availability of adsorbed water for plants [52], despite charcoal’s having a positive effect on soil water retention due to its high porosity [53].

Historically, charcoal production is perceived as one of the most intensive forms of forest exploitation in Europe, with the impacts of this procedure being permanently reflected in the soil’s chemical properties. In our study, while pH, CEC, TOC, TN, C/N and Al_o/Al_t levels in the kiln soil environment were all similar to levels in the control plots, base cation concentrations were substantially higher under the kiln platforms, with levels at around 60%–70% higher than those in control plots (Table 2). While Hardy et al. [12] also recorded higher CEC and C/N ratios, their other results differed somewhat from our own, possibly due to a different sampling design and/or topsoil stratigraphy. As with Hardy et al. [12], we recorded the depletion in exchangeable K; however, while these authors also recorded a depletion in the available P (attributing it to biological immobilization), we did not. In the same study, the authors noted that the soil’s chemical properties changed over time, following the abandonment of the kiln platform. In particular, they noticed that while the pH was close to neutral immediately after wood charring, it dropped gradually over time [54], due to base cation depletion from leaching and take-off, especially by plants. This suggests that it is not necessarily an automatic process; rather, we suggest that the soil chemistry of charcoal kiln platforms develops as an open system, with new balances being established according to the neighboring soil environment.

In addition to their impact on individual element concentrations, charcoal kilns can also play a substantial role in increasing the total available nutrient stocks, summed for 40 cm of soil depth (cf. Figure 4 and Table 4), with the presence of charcoal acting as an active sorbent in carbon-rich charred layers, resulting in significant increases [55]. According to some authors, this, together with increased CEC values, is the basis for the fertility and sustainability of the terra preta anthropogenic soils (anthrosol) of Amazonia, despite their having a totally different origin [40,56].

4.3. Charcoal-Rich Layer Biochemistry

Our results showed that depolymerization enzyme activity was significantly higher, and humification enzyme activity significantly lower, in O horizons compared with the rest of the topsoil (

Table 5. Results of the ANOVA tests for enzyme activity in the charcoal kiln (K) and control (C) plots. p-value significance: = 0.05–0.1, * = 0.01–0.1, ** = 0.001–0.01, *** = < 0.001, ns = non-significant. Depolenz = Depolymerization enzyme activity, Humifenz = Humification enzyme activity.

Enzyme (y ~)	Site	Horizon	p-Value
Depolenz	K	O + OM + OMC + MR + M	***
Depolenz	C	O + OM + M	***
Humifenz	K	O + OM + OMC + MR + M	***
Humifenz	C	O + OM + M	.
Depolenz	K	OM + OMC + MR + M	***
Humifenz	K	OM + OMC + MR + M	ns
β-Glucosidase	K	OM + OMC + MR + M	***
Chitinase	K	OM + OMC + MR + M	.
Leucine-aminopeptidase	K	OM + OMC + MR+M	***
Phosphatase	K	OM + OMC + MR+M	***
Arylsulphatase	K	OM + OMC + MR+M	*
Peroxidase	K	OM + OMC + MR+M	ns
Phenoloxidase	K	OM + OMC + MR+M	**

and Figure 6). At the kiln sites, in particular, values in the O horizons deviated so much that partial differences between the remaining horizons in the profile were obscured. Consequently, we reduced the dataset by removing the O horizon, allowing us to evaluate the OM and mineral horizons in greater detail. This follow-up analysis showed significant differences in the activity of the whole group of depolymerizing enzymes at kiln sites (Figure 7), while differences in the activity of humification enzymes turned out to be non-significant (although later tests confirmed a significant difference in phenoloxidase activity; see Figure 8d), although showing generally higher activity in the OMC horizons, with the activity levels being similar in the other horizons.

This same follow-up assessment (Figure 8) also confirmed increased activity in five extracellular enzymes that are involved in the transformation of C, N, P, and S (Table 3 and Table 5), an effect also noted by Lasota et al. [57]. Our results also confirmed an increase in β-glucosidase activity in the layer with charcoal (Figure 8a), along with increased phosphatase and, less strongly, chitinase activity in the OMC horizon. In addition, there was a significantly higher activity of arylsulfatase in the OMC horizon. These findings concur with those of Lasota et al. [26], who also reported higher levels in the OMC horizon.

Over the decades or centuries since platform abandonment, charcoal remnants will go through an aging process that alters the local biological, chemical, and/or physical properties [58]. If the charcoal kiln platform is relatively young (i.e., a short time has passed since its last use), it may also have a negative impact on soil enzyme activity, as demonstrated by Gómez-Luna et al. [59], who reported that forest soil arylsulfatase, acid phosphatase, β-glucosidase, and urease activities were inhibited by up to 70%, compared to a control plot. Similar outcomes have also been reported by Eivazi and Bayan [61].

In our study, we estimate the last use at 70 to 150 years previously; consequently, the revitalization of the soil environment means that the site can now be viewed as having ‘soil enriched with charcoal’ rather than as a former production platform with disturbed topsoil. Many studies have described the effect of charcoal on the microbiological properties of soils, including the activity of microbial communities and/or their composition, e.g., [12,59,60]. Other authors [61] have noted that the presence of biochar in the soil can increase P availability, thereby affecting bacterial communities and their extracellular enzyme activity. In our case, it can be assumed that this is primarily due to the activity of phosphatases in the OMC horizon (Figure 8b). Coomes et al. [62] also noted an increase in the amount of available P at charcoal platform sites; however, they recorded the highest available P at young kiln sites. Although P availability is expected to decrease over time as biological fixation progresses or with the P eventually precipitating into inaccessible forms through the process of chemisorption [63], the activity of mechanisms that make P available as a macrobioelement for soil microbiota and plants also tends to increase, generally resulting in a net increase in P availability.

In contrast, the reduced activity of humification enzymes, especially peroxidase, was somewhat surprising (Table 5). This may be an indication that this soil component is not equally suitable for the entire spectrum of the soil microbiota. As charcoal has a highly porous structure, it has a large internal surface area that provides a suitable environment for soil microorganisms, such as bacteria or mycorrhizal fungi [64]. Biochar, on the other hand, can have a less favorable effect on the development of mycorrhizal fungi, with subsequent effects on plant growth [11]. Warnock et al. [65], for example, noted the negative influence of charcoal on arbuscular mycorrhiza and ectomycorrhiza [66], whereby the formation of mycorrhizae was reduced and, subsequently, the availability of nutrients in the soil. Consequently, one can expect reduced activity in the soil enzymes that these fungi produce.

Ding et al. [67] also noted that the presence of charcoal in the soil environment can initiate changes in the structure and/or composition of the soil microbiota. Notably, this need not be due to any alteration in the source of available C for microorganisms; instead, it may be caused by changes in the availability of nutrients or in soil pH, in effect, adjusting the ecological niche in the soil environment to a different amplitude [26,68]. Carter et al. [69] also noted an increase in pH associated with the addition of charcoal, although they argued that it was probably of little ecological significance, as did Bonhage et al. [70], with the latter authors also recording an increase in the concentration of Ca, K, and Mg and increased TN and OC stocks. In our own work, we recorded an increase in POxC C-stock in the OMC layer. While POxC values in the lower mineral horizon were somewhat higher than those in the control (Table 3), levels in the kiln OM layer remained similar to those in the control (Figure 8e). This may reflect the absorption capabilities of charcoal, whereby labile forms of soil carbon are absorbed and are subsequently used by microorganisms as a substrate for growth [69,71,72].

5. Conclusions

In this study, we investigated the hydrophysical, chemical, and biochemical soil properties of charcoal kiln production platforms, and compared these with those of neighboring forest soils. In doing so, our aim was to detect any differences in individual soil horizons that could be attributed specifically to the anthropogenic soil environment of kilns, as opposed to forest soil control plots.

At kiln sites, both soil coloration and the specific features related to nutrient concentrations, with the localized accumulation of C and N, and, to a lesser degree, P and Ca, represent clear anomalies compared with neighboring forest soils. The chemical and mechanical changes observed in kiln horizons led to increased soil microbiota activity and plant root abundance, with significantly higher rooting densities at kiln sites, resulting from both increased nutrient availability and an increase in pore distribution, the high abundance of capillary pores increasing soil aeration and water-holding capacity.

Examination of the soil’s biochemical properties indicated significantly higher depolymerization enzyme activity and a significantly lower humification enzyme activity in kiln O horizons. Examination of soil stratigraphy without O horizons enabled us to recognize high leucine-aminopeptidase, phosphatase, and arylsulphatase and phenoloxidase activities in the charcoal-rich horizons, along with low humification enzyme activity (especially peroxidase) in horizons with less processable carbon. Thus, charcoal-rich horizons can be perceived as containing a highly effective sorbent on a mechanical base, with a highly recalcitrant structure that captures soil particles and interacts with the surrounding environment. After long periods, these carbon-rich kiln soil layers come to represent a soil microenvironment with high biological activity and soil biota concentration, linked with high root abundance.

Our results indicate that long-abandoned charcoal kiln platforms represent hotspots for plant and soil microbiota growth, thanks to the substantial changes in physical and chemical soil properties compared with neighboring control sites. In future studies, we will examine microbial nutrient cycles in more detail, particularly in relation to the precise dating of kiln platform age, which will help in estimating the time needed for re-establishing nutrient and biological balance at similar sites.