Renaturalization of Ex-Arable Arenosols: Phytocenosis Development and the Dynamics of Sandy Soil Properties

Abstract

The abandonment of agricultural land has strong implications for the environment and societal wellbeing. Combining field monitoring data with a detailed analysis of the existing literature, we explored the potential factors impacting the variability of annual/biennial and perennial plant species. We identified plants covering sandy Arenosols under agricultural land use for at least 50 years as of 1995 in order to define a strategy for their restoration. The results of the study distinguished 81 different plant species and 23 botanical families spread over 27 years of land abandonment. The most dominant families were Asteraceae, Fabaceae, and Poaceae. The productivity of the abandoned land’s phytocenosis increased as the duration increased (r = 0.70). A positive relationship was established between the phytocenosis biomass and the number of cereal grass species (r = 0.44). The abandonment of the land had positive effects on soil fertility: the SOC concentration in the Ah horizon increased from 9.9 ± 0.08 to 14.5 ± 1.17 g kg⁻¹, the Ah thickness increased by 3 cm, and the SOC stocks increased by 24.51 Mg ha⁻¹. These results will allow us to study the vegetation dynamics in abandoned land and will serve as a basis for the selection of appropriate species in eventual restoration projects.

1. Introduction

Sandy soils are naturally infertile and have low organic matter contents, so the importance of conserving or improving such soils is enshrined in legislation, such as the National Climate Change Management Policy and the National Environmental Protection Strategy. The Kyoto Protocol emphasizes that soil is one of the main CO₂ reserves, which must be protected and, if possible, increased. Additionally, the General Union Environment Action Programme to 2030 aims at protecting, preserving, and restoring marine and terrestrial biodiversity and the biodiversity of inland waters inside and outside protected areas by, inter alia, halting and reversing biodiversity losses, improving the state of ecosystems and the functions and services that they provide, improving the state of the environment—in particular air, water, and soil—and combating desertification and soil degradation. Nevertheless, low-productivity coarse-textured soils are risky for the cultivation of agricultural products. Therefore, the exploration of the rational use of resources in such soils requires special attention. In this way, abandoned farmlands can provide valuable opportunities for nature restoration together with the EU Biodiversity Strategy for 2030 and the Post-2020 Global Biodiversity Framework for biodiversity protection and ecosystem restoration [1]. The return to agricultural use also resonates with the European Green Deal’s Farm to Fork Strategy by providing options for sustainable food production and economic returns for farmers.

One of the possible approaches is the formation of abandoned lands or arable soil left to rest on such soils. Agricultural land abandonment is a global land-cover change that has increasingly strong implications for the environment (e.g., biodiversity, carbon sequestration, novel ecosystems, wildfires) and societal wellbeing (e.g., livelihood, agricultural landscapes) [2]. Agricultural land abandonment contributes to the accumulation of organic matter in the soil, which improves the physical and chemical properties of the soil and reduces the negative impact of climate change [3,4,5]. In addition to preventing soil degradation, fallows also contribute to the conservation of plant biodiversity. Under certain conditions, abandoned lands may experience a reduction in plant species diversity [6].

Along with the development of abandoned lands, the expansion of ecological farms with elements of natural landscapes and biodiversity remains relevant. The European Union provides payments for well-managed natural grasslands and meadows, and one of their objectives for the future is to preserve landscapes and biodiversity in order to promote the development of natural grasslands. Investigation of the structure of perennial grasslands and phytocenosis succession process, as well as of changes in perennial meadows and pastures, is relevant not only from the economic and social but also from the scientific point of view [7]. For political decisions, it is important that biodiversity issues and values are mainstreamed, and that decision-making is based on scientific data.

There is considerable interest in being able to predict the dynamics of plant populations on abandoned land. For many species, long-term variability may be determined by predictable factors. However, evidence for the role and nature of such variability may be hard to obtain, and there is a clear need for more long-term time series on plant species distributed in abandoned lands. In the absence of such data, however, it is possible, by combining the field monitoring data with detailed analysis of the existing literature, to try to explore the potential factors impacting the variability of annual/biennial and perennial plant species. In this way, the longevity of plants’ seeds is important. The actual longevity of seeds in the soil depends on an interaction of many factors, including the intrinsic dormancy of the seed population, depth of seed burial, frequency of disturbance, environmental conditions (e.g., light, moisture, temperature), and biological processes, such as predation, allelopathy, and microbial attack [8,9]. Thus, abandoned land successions always undergo a cluster of diverse processes affected by many factors, and the results may differ in each case. Understanding the dynamics of natural spread succession will help us to better understand both the final and intermediate stages [10].

The succession of ex-agrogenic vegetation can manifest both in a directional change in the species composition of the vegetation and in the formation of plant matter stocks [11]. In post-agrogenic agrocenoses, the patterns of formation and mineralization of aboveground phytomass and mortmass have a significant effect on the formation of SOC [12]. When the phytomass dies away, mortmass is formed in middle-aged fallows, a significant part of which is litter. The low concentration of SOC in arable soils is due to the amount and quality of plant residues entering the soil, the removal of part of the plant production with the crop, changes in the conditions of humification and mineralization during soil cultivation, and erosion losses.

The aims of this study were to determine changes in the species composition of the phytocenosis on sandy soils (Endocalcaric Arenosols) in abandoned land, to determine the dominant species, and to identify patterns of phytocenosis biomass formation and its influence on soil properties.

2. Materials and Methods

2.1. Experimental Site Description

The long-term experiment (beginning in 1995) was conducted at the Lithuanian Research Centre for Agriculture and Forestry (54°33′49.8″ N 25°05′12.9″ E, Lithuania, East Baltic region, Central Europe) (Figure 1). From the geomorphological point of view, the survey site was located in an area where the surface consists of binary rocks: fluvioglacial and Gruda-stage washed moraine sandstone. The relief is a slightly undulating plain.

The soil was formed on fluvioglacial deposits, and it had the following profile: Ah-AB-B1-B2-1C-C2 [13]. The texture of the soil was sandy, comprising ~88% sand, ~8% silt, and ~4% clay in the top, and ~9% silt and ~5% clay in the subsoil [14] (

Table 1. Soil characteristics of the experimental site.

Horizon	Depth	Texture
		Sand	Silt	Clay
	(cm)	%
Ah	0–28	88.58	7.67	3.75
AB	28–43	86.89	8.66	4.45
B1	43–70	84.55	7.95	7.50
B2	70–96	86.80	4.55	8.65
C1	96–107	91.43	3.53	5.04
C2	107–121	98.42	1.16	0.42

). According to soil texture and the structure of the profile, the soil belonged to the Arenosol group (Endocalcaric Arenosol). The upper layer of the soil was relatively low in organic carbon, the amount of which was significantly reduced in the AB and B1 horizons. The soil reaction was weakly acid (pH_KCl 6.0–6.1). The concentration of available phosphorus at the beginning of the experiment was 157 mg P₂O₅ kg⁻¹, while the concentration of available potassium was 170 mg K₂O kg⁻¹.

According to B. Alisov’s classification, the location of the experiment was in a temperate climate zone and the subregion of Atlantic–European continental mixed and broadleaved forests [15]. This region is characterized by a moderate climate, with a mean long-term (1981–2010) annual precipitation of 685 mm and an annual mean air temperature of 6.7 °C (the standard climate norm (SCN)) [16]. These hydrothermal conditions are conducive to the precipitation, filtration, and leaching of biogenic elements from the upper soil layers.

2.2. Treatment and Experimental Design

Our research on changes in plant succession after the cessation of agricultural activity was part of an experiment initiated in 1995 that investigated changes in soil properties after the conversion of arable land to other types of land use. Up to 1995, this soil was considered to be arable. The total size of each land-use site was 400 m² (20 m × 20 m). The land-use conversion methodology that we used is currently the most recommended methodology to aid in decision-making on environmental issues in Europe and the USA [17,18]. No soil working was practiced in the abandoned land, but sporadic woodcutting was performed as needed in such a way as to avoid overgrowth of trees (self-afforestation). During the study period, a natural vegetation phytocenosis typical of sandy soils in this region was formed on the site of abandoned ex-arable land.

2.3. Soil Sampling and Analytical Methods

Soil samples for the determination of chemical indicators (Ah horizon) were taken before the beginning of the experiment in 1995, as well as in 2022, in three replicates. The soil samples for the analytical determinations were homogenized and air-dried before being gently crushed and passed through a 2 mm mesh sieve. The soil bulk density (Ah horizon) was determined by the core method, using a metal ring pressed into the soil (intact core) and determining the weight after drying [19]. Core cutter samples were taken in three replicates from the upper surface layer (Ah horizon).

The soil chemical properties were determined as follows:

pH_KCl: by the ISO 10390:2005 potentiometric method (1 mol l⁻¹ KCl using a soil/solution ratio as 1:2.5).

C_org: by the Duma method (after dry combustion), ISO 10694:1999.

Plant available P₂O₅ and K₂O were extracted using 0.03 M ammonium lactate (Egner–Riehm–Domingo (A–L) method).

Soil organic carbon stocks in the A horizon were calculated as follows [20]:

SOC_stock (Mg ha⁻¹) = SOC_con × BD × depth ÷ 10

(1)

where SOC_con is the soil organic carbon concentration (g kg⁻¹), BD is the bulk density of the A horizon (Mg m⁻³), depth is the thickness of the humic A horizon layer (cm), and 10 is the coefficient used to calculate the SOC stocks in Mg ha⁻¹.

2.4. Determination of the Species Composition and Biomass of Plants

The analysis of the species composition of plants in the experimental site was carried out in 1995, 2000, 2004, 2015, and 2022. The vegetation cover was estimated according to Braun-Blanquet’s method [21]. Identification of plant species was carried out using the K.K. Vilkonis atlas [22].

For the assessment of changes in plant succession, they were systematized according to growth duration (i.e., annual, biennial, perennial) and classified by plant family. The species composition of the plants was determined in 4 replicates in an area of 0.25 m².

Plant biomass (DM yield Mg ha⁻¹) was determined annually after the flowering of the dominant plant species (second decade of July) from an area of 0.25 m² with 4 replicates. For the determination of dry matter, the plant samples were dried at 105 °C.

2.5. Data Analyses and Statistics

The data were structured and analyzed using the Microsoft Excel software package.

Standard error (SE) values were used to estimate the deviations of the soil chemical parameters from the mean values. A leaner regression analysis was used to reveal the relationship between biomass DM yield and the duration of the experiment, between biomass DM yield and the amount of precipitation, and between biomass DM yield and the number of plant species.

3. Results

3.1. Diversity of Plant Species

Under the scheme of the experiment, the formerly arable lands were transformed by leaving them abandoned to become natural grasslands. In this way, soil renaturalization took place in a spontaneous manner; however, this influenced the soil properties and plant rotation in the naturally formed grasslands.

Various plant species grow in different communities, taking up respective niches therein. Under changing environmental conditions, the existence of various species in these communities changes accordingly, leading to alterations in the floristic composition of communities, i.e., some species disappear while others appear. The zone of all herbaceous species formed as a result of the abandonment of the arable land is presented in

Table 2. Plant species examined on abandoned sites of Endocalcaric Arenosol over a 27-year period, with frequency (1–6–10–20–27).

Family Name	Latin Name of Species	Life Type	Abandonment Time
			1 Year	6 Years	10 Years	20 Years	27 Years
Amaranthaceae	Amaranthus retroflexus L.	Y	r
Apiaceae	Pimpinella saxifraga L.	P			r	+	r
Asteraceae	Achillea millefolium L.	P	+	r	1	1	1
	Artemisia vulgaris L.	P	+	+	1	r
	Carduus acanthoides L.	B	r	r			1
	Carlina vulgaris L.	P/B			+
	Centaurea cyanus L.	Y	+
	Cirsium vulgare L.	P			+
	Conyza canadensis L.	Y		+	1
	Erigeron acris L.	B		r	+
	Filago arvensis L.	Y	r
	Gnaphalium silvaticum L.	P		r	+
	Helichrysum arenarium L.	P			+
	Hieracium vulgatum L.	P				1
	Pilosella officinarum F. W. Schultz et Sch. Bip.	P			1	2	3
	Phalacroloma annuum L.	P/B					r
	Senecio jacobaea L.	P/B			+
	Solidago virgaurea L.	P		r			r
	Taraxacum officinale L.	P		r	+		r
	Thlaspi arvense L.	Y	r
	Tripleurospermum inodorum L.	Y/B	1
Boraginaceae	Echium vulgare L.	B			+
	Myosotis micrantha Pall. ex Lehm.	P	r
Brassicaceae	Berteroa incana L.	B	r
	Camelina microcarpa L.	Y	r
	Capsella bursa pastoralis (L.) Medik.	Y	r
	Erophyla verna L.	Y	+
	Erucastrum gallicum L.	Y/B				+
	Raphanus raphanistrurm L.	Y				r
Campanulaceae	Jasione montana L.	B			+
	Campanula patula L.	P				r	+
	Campanula glomerata L.	P					r
Caryophyllaceae	Arenaria serpyllifolia L.	Y	+	+
	Cerastium arvense L.	P			+		r
	Cerastium holosteoides L.	P	r
	Psammophiliella muralis L.	Y
	Scleranthus annuus L.	Y	1
	Silene pratensis L.	P	+	+	+
	Silene vulgaris L.	P	+	+	+
	Spergula arvensis L.	Y	+
	Spergularia rubra (L.) J. Pressl et C. Presl.	P/Y/B
	Stellaria holostea L.	P					1
Chenopodiaceae	Chenopodium album L.	Y	+
Convolvulaceae	Convolvulus arvensis L.	P			+	+	+
Equisetaceae	Equisetum arvense L.	P	+	+	+	2
Fabaceae	Lotus corniculatus L.	P		1	2	+
	Medicago falcata L.	P				1	1
	Medicago lupulina L.	P/Y					1
	Melilotus albus Medik.	P/Y	r
	Trifolium arvense L.	Y	r	r	+		1
	Trifolium repens L.	P	r				+
	Vicia angustifolia L.	Y		r
	Vicia cracca L.	P	r	r	+		1
	Vicia lathyroides L.	Y				+
Fumariaceae	Fumaria officinalis L.	Y				1
Geraniaceae	Erodium cicutarium L.	Y	r
Globulariaceae	Globularia punctat Lapeyr.	P					r
Hypericaceae	Hypericum perforatum L.	P	r	+	1		r
Illecebraceae	Herniaria glabra L.	P	r
Poaceae	Agrostis gigantea L.	P			r
	Bromus inermis L.	P			r
	Dactylis glomerata L.	P				1	+
	Elytrigia repens L.	P	3	3	3		r
	Festuca arundinacea Schreb.	P				4	3
	Festuca ovina L.	P					2
	Festuca rubra L.	P			+		r
	Phleum pratense L.	P		r	+
	Secale cereale L.	Y	+
	Setaria viridis L.	Y	+	r			r
Polygonaceae	Polygonum aviculare L.	Y	+
	Rumex acetosella L.	P		+
	Rumex obtusifolius L.	P				+
Plantaginaceae	Plantago lanceolata L.	P	r	r	+
	Plantago major L.	P	r
Ranunculaceae	Consolida regalis Gray.	Y	+		+
Rosaceae	Fragaria vesca L.	P				r
	Geum urbanum L.	P			+
	Potentilla argentea L.	P	1	r	+
Scrophulariaceae	Linaria vulgaris L.	P	r	r	+
	Veronica arvensis L.	Y	r
Violaceale	Viola arvensis Murray.	Y	1	r

Note—Life type: Y = one-year herb; B = biennial herb; P = perennial herb. r = <5% very few individuals; + = <5% few individuals; 1 = <5% numerous individuals; 2 = 5–25%; 3 = 25–50%; 4 = 50–75%; 5 = 75–100%.

.

In 1995, high species diversity was established, including 39 plant species belonging to 17 families; 5 species (1–3 score according to the Braun-Blanquet scale) were identified as dominant. It was revealed that ex-arable land’s seed reserves affect the abandoned site’s vegetation. The abandoned land was located in arable soil that, over a long period of time, had accumulated weed seed stocks typical of sandy soil, including mayweed (Tripleurospermum inodorum L.), pennycress (Thlaspi arvense L.), pansy (Viola arvensis Murray.), cornflower (Centaurea cyanus L.), shepherd’s purse (Capsella bursa pastoris (L.) Medik.), goosefoot (Chenopodium album L.), amaranth (Amaranthus retroflexus L.), foxtail (Setaria viridis L.), and couch grass (Elytrigia repens L.). After 1 year of abandonment, Elytrigia repens of the Poaceae family was the dominant species in the community (3 score according to the Braun-Blanquet scale).

Despite the spread of the aforementioned plants in the first year of the arable land’s abandonment, a significant part of the annual phytocenoses disappeared as the renaturalization period lengthened. After performing a botanical analysis of the composition of the soil plants, after 6 years of renaturalization (in 2000), a decrease in plant species was detected (i.e., 25 plant species) compared to their quantity at the beginning of the abandonment. These 25 species belonged to 10 families. The composition of annual and biennial phytocenoses of different species in 2000 was ~32%, while in 1995 it was ~49%. Conversely, after several years of abandonment, the perennial plants species increased by 24%. Even the annual or biennial plants held on in the naturally occurring grassland and were still found after 6 years of abandonment. Such annual plants included thyme-leaf sandwort (Arenaria serpyllifolia L.), foxtail (Setaria viridis L.), hare’s-foot clover (Trifolium arvense L.), and field pansy (Viola arvensis Murray.), while the biennial plants were dominated by plumeless thistle (Carduus acanthoides L.). There were newly grown annual grasses, such as narrow-leaved vetch (Vicia angustifolia L.) and biennial bitter fleabane (Erigeron acris L.). Perennial couch grass (Elytrigia repens L.) grew steadily until 2000. During the period of 6 years of abandonment, Elytrigia repens L. remained the dominant species in the community, with a coverage score of 3 according to the Braun-Blanquet scale. Additionally, in 2000, bird’s-foot trefoil (Lotus corniculatus L.) had established itself in the soil, with a coverage score of 1 according to the Braun-Blanquet scale. Unlike Elytrigia repens L., Lotus corniculatus L. is a perennial grass cultivated by farmers as a high-protein fodder plant.

The following plants with a short life cycle were not found in our study after 6 years of abandonment: amaranth (Amaranthus retroflexus L.), hoary alyssum (Berteroa incana L.), little-pod false flax (Camelina microcarpa Andrz. subsp. pilosa (DC.) Jav.), shepherd’s purse (Capsella bursa pastoralis (L.) Medik.), common stork’s-bill (Erodium cicutarium L. Her.), honey clover (Melilotus albus Medik.), blue scorpion grass (Myosotis micrantha Pall. ex Lehm.), broadleaf plantain (Plantago major L.), pennycress (Thlaspi arvense L.), corn speedwell (Veronica arvensis L.), cornflower (Centaurea cyanus L.), field larkspur (Consolida regalis Gray.), smartweed (Polygonum aviculare L.), rye (Secale cereale L.), and corn spurry (Spergula arvensis L.).

In 2004, high species diversity remained, including 33 plant species. These plant species belonged to 14 families; 8 species (1–3 score according to the Braun-Blanquet scale) were identified as dominant. It was revealed that the ex-arable land’s plant species after 10 years of abandonment survived at the same quantity as observed at the beginning of the experiment. It was noticeable that the abandoned ex-arable land’s vegetation began to be dominated by three families of plant over the first decade: Asteraceae, Fabaceae, and Poaceae. After 10 years of abandonment, perennial plants species dominated, accounting for 79% of all plant species.

Discussing individual plant species over this decade, perennial couch grass (Elytrigia repens L.) grew and its coverage (3 score according to the Braun-Blanquet scale) did not change compared with its coverage in 2000. Moreover, by the end of this decade, plants, such as common yarrow (Achillea millefolium L.), mugwort (Artemisia vulgaris L.), Canadian horseweed Conyza canadensis L., and mouse-ear hawkweed (Pilosella officinarum F. W. Schultz et Sch. Bip.) (1 score according to the Braun-Blanquet scale), appeared. In 2000, we found that bird’s-foot trefoil (Lotus corniculatus L.) began to spread more, and its coverage score was 2 according to the Braun-Blanquet scale.

In 2015, 18 plant species belonging to 11 families were found in the natural grassland, created as a result of a long-term absence of tillage, among which the predominant species were common yarrow (Achillea millefolium L.), mouse-ear hawkweed (Pilosella officinarum F. W. Schultz et Sch. Bip.), tall fescue (Festuca arundinacea Schreb.), orchard grass (Dactylis glomerata L.), mugwort (Artemisia vulgaris L.), and field horsetail (Equisetum arvense L.). Due to the growth of perennial cultivated grasses, such as tall fescue (Festuca arundinacea Schreb.) (4 score according to the Braun-Blanquet scale) and orchard grass (Dactylis glomerata L.) (1 score according to the Braun-Blanquet scale), annual grasses were much more dominant in 2015. The aforementioned cultivated grasses could have been sown from nearby fields.

Depending on the plant species that are naturally adapted to the local conditions, established, and competing, the succession can develop in very different directions and at different speeds. Meanwhile, after 20 years of abandonment, perennial couch grass (Elytrigia repens L.) disappeared, or its coverage dropped significantly. The coverage of bird’s-foot trefoil (Lotus corniculatus L.) also dropped.

The composition of annual and biennial phytocenoses of different species after 20 years of abandonment was ~22%, while in 2000 it was ~32%, and in 1995 it was ~49%. On the other hand, after several years of abandonment, the perennial plant species increased by 10% in comparison with 2000 and 27% in comparison with 1995.

Yarrow (Achillea millefolium L.) (1 score according to the Braun-Blanquet scale) and mouse-ear hawkweed (Pilosella officinarum F. W. Schultz et Sch. Bip.) (2 score according to the Braun-Blanquet scale) were perennial species that showed continuous spread. These small herbaceous plant communities usually settle in grasslands, pine forest sites, and woodlands. Suitable conditions for the succession of yarrows and mouse-ear hawkweed were created by the woody vegetation that began to spread on the natural grassland site.

Throughout the entire study period (1995–2015), mugwort (Artemisia vulgaris L.), which is very widespread in Lithuania, was continuously detected, although its prevalence was not high. In arable land, this plant is considered to be a difficult-to-control weed, as it is an allelopathic plant that frequently hampers the germination and growth of crops.

In 2022, it was revealed that three families of plant continuously dominated the vegetation in the abandoned ex-arable land over the nearly three decades of the study: Asteraceae, Fabaceae, and Poaceae. This dominance was the same as in 2004 and 1995.

The order of species abundance during the 27 years of land abandonment was as follows: 20 years (18 species, 11 families) < 27 years (25 species, 9 families) < 6 years (25 species, 10 families) < 10 years (33 species, 14 families) < 1 year (39 species, 17 families), showing the trend of undulation in plant species quantity.

The main contribution to the formation of the total stock of aboveground plant matter was made by cereals, such as tall fescue (Festuca arundinacea Schreb.) and sheep fescue (Festuca ovina L.). Species of the Poaceae family occurring individually but forming a significant amount of phytomass—such as Festuca arundinacea Schreb. and Festuca ovina L., together with the less common orchard grass (Dactylis glomerata L.), red fescue (Festuca rubra L.), and foxtail (Setaria viridis L.)—contributed considerably to the total reserve of plant matter.

The share of participation of other species, such as mouse-ear hawkweed (Pilosella officinarum F. W. Schultz et Sch. Bip.), in the formation of the total phytomass stock was less, despite their considerable coverage. This species of the Asteraceae family began to spread after 10 years of abandonment. Due to its invasive and allelopathic characteristics, its coverage changed from a score of 1 according to the Braun-Blanquet scale in 2004 to a score of 2 in 2015 and 3 in 2022. Additionally, common yarrow (Achillea millefolium L.) and plumeless thistle (Carduus acanthoides L.) from the Asteraceae family grew in coverage, with a score of 1 according to the Braun-Blanquet scale.

The Fabaceae family was the third most abundant in the abandoned land after 27 years. Fabaceae plants, such as Sickle medick (Medicago falcata L.), shamrock (Medicago lupulina L.), hare’s-foot clover (Trifolium arvense L.), and tufted vetch (Vicia cracca L.), grew in abundance. Hare’s-foot clover (Trifolium arvense L.) and tufted vetch (Vicia cracca L.) began to grow in 1995 and were found after 10 years of abandonment; later, these species were lost, and their growth was found again after 27 years. Shamrock (Medicago lupulina L.) was found on the abandoned land only in 2022. Sickle medick (Medicago falcata L.) began to spread in the grassland only after 20 years of abandonment. However, all of these species had a coverage score of only 1 on the abandoned land plot according to the Braun-Blanquet scale.

Scots pine (Pinus sylvestris L.) and common bird cherry (Padus avium Mill.) were recorded on the abandoned site since 2002 and 2015, respectively, as the woody vegetation started to grow and form a young forest. The Scots pine (Pinus sylvestris L.) seeds had traveled from a nearby pine-afforested site. To avoid the growth of the trees, they were removed from the soil over time.

3.2. Plant Biomass and Soil Properties as Related Factors Together with Plant Species

During the research period, the mean soil vegetation biomass yield was 1.97 Mg ha⁻¹, with a median of 1.77 Mg ha⁻¹, and varied from 0.9 to 4.21 Mg DM ha⁻¹ depending on environmental factors (Figure 2). Approximately 15–17 years after the abandonment of the arable land, a trend of increasing plant biomass yield could be observed. This could have been influenced by several factors, including changes in the botanical composition of the soil vegetation, climatic changes, and an increase in organic carbon accumulation in the soil. According to the correlation analysis, it was determined that the biomass yield was not dependent on the annual amount of precipitation (r = 0.001).

During the 1995–2022 period, a slight tendency of increasing precipitation was determined in the territory of Lithuania; however, the amount of annual precipitation decreased slightly in the most recent decades. The higher biomass yield could have been influenced by the increase in air temperature during the plant vegetation period (March–September), as well as in the months of October–December, which created better conditions for the plants’ wintering.

An analysis of changes in the botanical composition of the soil vegetation showed that the yield of plant biomass partially depended on the number of plant species of the Poaceae family (r = 0.44).

The increase in organic carbon accumulation in the humic horizon could also have an influence on the biomass yield (

Table 3. Soil properties of the abandoned land.

	Ah Depth cm	Soil Bulk Density	pHKCl	Available P mg kg−1	Available K mg kg−1	SOC g kg−1	SOC Mg ha−1
1995	28.0	1.33 g ha K	6.0 ± 0.08	157 ± 4.9	170 ± 4.7	9.9 ± 0.08	36.8 ± 0.86
2022	31.0	1.37 8 ha K	6.1 ± 0.11	131 ± 7.3	123 ± 8.7	14.5 ± 1.17	61.4 ± 5.00

). During the research period, the concentration of organic carbon in the Ah horizon increased from 9.9 ± 0.08 to 14.5 ± 1.17 g kg⁻¹, the thickness of the Ah horizon increased by 3 cm, and the accumulation of organic carbon increased by 24.51 Mg ha⁻¹. The sequestration of organic carbon in the abandoned land averaged 0.907 Mg ha⁻¹ per year. Even though plant biomass remains on the soil surface after the end of vegetation, the concentration of mobile nutrients in the Ah horizon decreased in the presence of excess moisture. The available phosphorus decreased by 26 mg P₂O₅ kg⁻¹, while the available potassium decreased by 47 mg K₂O kg⁻¹, since it is easier to leach from the upper soil layer in sandy soils. The decrease in the concentration of available potassium was not critical in limiting the formation of natural vegetation biomass. The soil bulk density after 27 years of abandonment changed from 1.33 iki 1.37 Mg m⁻³.

4. Discussion

4.1. Plant Richness

Without management, after abandonment, the vegetation returns to a natural basis, providing carbon capture, biodiversity recovery, and recreational opportunities [1]. Succession can also support spontaneous restoration at low cost [23]. Furthermore, it is always a possibility to return to agricultural land use for sustainable food production and economic returns for farmers.

In the temperate climate zone, with a sufficient amount of precipitation in the first year of abandonment, plant species typical of arable soil were preserved in the phytocenosis, due to the presence of seeds of these plants in the soil. Similar processes of development of successions are described in the works of other researchers. The authors of [24] analyzed the longevity of different weed species and found that, regardless of management practice in arable lands, some weed seeds can remain dominant in the soil for ~20 years. Such species included shepherd’s purse (Capsella bursa pastoris (L.) Medik.), goosefoot (Chenopodium album L.), smartweed (Polygonum aviculare L.), and corn spurry (Spergula arvensis L.), which were also identified in our abandoned land.

Popular soil plants, such as pennycress (Thlaspi arvense L.), goosefoot (Chenopodium album L.), and couch grass (Elytrigia repens L.), have also been identified by Slovak scientists [25] in sandy soils in semi-natural grassland successions (formed as a result of the abandonment of arable lands). The large variety of such plants at the beginning of soil renaturalization was because the natural grassland was planted in soil that was arable for many years. Despite farming in Lithuania’s climatic zone being carried out over a long period of time, it concentrates weed seed reserves characteristic of arable sandy topsoil.

Research by other scientists [26] has shown that plants of the Poaceae family often exhibit allelopathic properties. In this way, Poaceae plants affect one another and influence the survival of other plants in grasslands. According to other [27] studies on couch grass (Elymus repens) and Timothy grass (Phleum pratense L.), it has been proven that the pollen of Phleum pratense L. has a negative effect on couch grass (Elymus repens). In this way, there is a possibility of the reduction or complete disappearance of these plants in natural grasslands. This partially explains why some plants, such as couch grass (Elytrigia repens L.), were still found in the natural grassland in the first 10 years of abandonment (until 2004), but more than 2 decades later they were no longer detected. Despite the strong roots of couch grass (Elytrigia repens L.), its extinction could have been caused by the competing species Timothy grass (Phleum pratense L.). In [24], we can find another reason for the disappearance of Elytrigia repens L.—the longevity of Elytrigia repens L. seeds in the soil can be ~4–6 years.

According to research conducted in Latvia [28], yarrow (Achillea millefolium L.) and couch grass (Elymus repens L.) covered a considerable area of ex-arable land during the renaturalization of soil. At the beginning of the plants’ establishment, yarrow (Achillea millefolium L.) and mouse-ear hawkweed (Pilosella officinarum F. W. Schultz et Sch. Bip.) almost always occupied small areas, often at the edges of other grassland communities or at the edges of the forests, but after some time they started to spread and cover very large areas of grassland [29]. These kinds of plants may have contributed to the overall decline in species in 2004, as they grow like a dense carpet and prevent the occurrence of other plant species in the grasslands. Plants, such as yarrow (Achillea millefolium L.) and mouse-ear hawkweed (Pilosella officinarum F. W. Schultz et Sch. Bip.), may also increase their density and start to crowd out other existing natural grassland species. This trend became more evident with the growth of mouse-ear hawkweed (Pilosella officinarum F. W. Schultz et Sch. Bip.). In 2004, Pilosella officinarum F. W. Schultz et Sch. Bip. had a coverage score of 1 according to the Braun-Blanquet scale, increasing to a score of 2 in 2015 and 3 in 2022.

Mouse-ear hawkweed (Pilosella officinarum F. W. Schultz et Sch. Bip.) grows close to the ground in a rosette pattern that forms dense mats [30]. It is native to Britain, Europe, and Asia, but now occurs as a serious weed in Australia, New Zealand, Canada, Argentina, Chile, and the USA [31,32]. It is an undesirable invader on account of its vigorous growth due to stolon production and wind-dispersed seeds. Pilosella officinarum displaces the inter-tussock vegetation, leading to loss of forage and biodiversity. It is also allelopathic, meaning that it prevents the germination and growth of other plants by producing biochemicals and secreting them into the surrounding soil [33].

It has been found that some aggressive and competing plant species can temporarily stop the development of succession. Other scientists [34] have indicated that tall fescue (Festuca arundinacea Schreb.) is a typical and commonly found grassland plant in Central and Southeastern Europe, among other regions. According to biodiversity studies conducted in Estonia [35], it was found that tall fescue (Festuca arundinacea Schreb.) spreads in areas where trees and bushes are cut down, which was the case in the investigated natural grassland phytocenosis. In addition, these plants can negatively affect the germination and growth of different plants. According to studies by other authors [36,37], tall fescue (Festuca arundinacea Schreb.) and orchard grass (Dactylis glomerata L.) show high levels of allelopathic properties compared with other plants.

The order of species abundance over 27 years of land abandonment decreased from 39 species to 25 species after 6 years, then increased to 33 species after 10 years, before again decreasing to 18 species after 20 years, and then increasing to 25 plant species after 27 years. In the early stage of abandonment, the species diversity was high because many opportunists began to invade [38]. In this way, over the 3 decades of abandonment, the plant richness changed in an up-and-down manner.

4.2. Plant Biomass and Richness-Driving Factors

With the increasing duration of the abandonment, the productivity of the phytocenosis increased. The same pattern has been described in scientific work by other authors [39], who also emphasized that species richness was linearly correlated with aboveground biomass, whereas plant abundance showed a humpbacked relationship with aboveground biomass across all successional stages. In different soil and climatic conditions (i.e., the Kostanay Region of northern Kazakhstan), a decrease in the productivity of the fallow with increasing stages was established [40]. We assume that in our experiment, the increase in biomass yield with increasing duration of abandonment could have been caused by several factors, including changes in the botanical composition of the soil vegetation, climatic changes, and an increase in organic carbon accumulation in the soil. Similar results are presented in another scientific study [41], which found that aboveground biomass was highly affected by environmental driving factors (e.g., soil properties, plant traits, topographic properties) and that plant height had a decisive influence on biomass yield. Correlation analysis of changes in yield revealed a moderate relationship (r = 0.44) of its value with the number of species of Poaceae family grasses in our experiment. No relationship was established between rainfall and aboveground biomass yield. We assumed that the increase in biomass yield may be related to the lengthening of the growing season in Lithuania, since over the past 30 years the average annual air temperature has increased by 1.0 °C [42], and the longer growing season allows plants to form more abundant biomass. However, this assumption requires additional research.

The positive effect of plant abundance on the accumulation of organic carbon in the soil has been established. Many researchers note that, in the initial years, the accumulation of organic carbon occurs slowly, and then increases significantly after 20–30 years [43,44,45]. In Arenosol, for 27 years, the concentration of organic carbon in the Ah horizon increased by 48%, and the thickness of the horizon increased by 3 cm, as a result of which the organic carbon reserves increased by 66.5%. Soil carbon storage reduces carbon dioxide emissions; thus, abandoned agricultural fields represent a valid opportunity to mitigate emissions from the agriculture sector [46,47]. However, the conversion of arable soil to abandoned land did not provide a stable concentration of nutrients in the soil. Over 27 years, the concentration of available phosphorus in the Ah horizon decreased by 26 mg P₂O₅ kg⁻¹, while that of potassium decreased more significantly, by 47 mg K₂O kg⁻¹. These losses were due to the hydrothermal regime in the territory of Lithuania. During the year, depending on the soil texture, 35–40% of atmospheric precipitation infiltrates, along with which nutrients are leached from the upper soil horizon [48]. Under other hydrothermal conditions, the leaching regime of soils can lead to a decrease in soil acidity, but the content of nutrients remains at a high level [49]. In regions with less rainfall, abandoned soils may experience an accumulation of nutrients from the topsoil [50,51]. The soil bulk density after 27 years of abandonment was observed to not change significantly and was typical of sandy soil. As the abandonment period increases along with the amount of soil organic carbon, the soil bulk density may decrease [52,53,54]. Consequently, the processes occurring in ex-agrogenic soil may have a different direction, depending on the soil and climatic conditions and the stage of the abandoned land.

Long-term investigation of ex-agrogenic Arenosol conversion in other land-use types in the temperate climate zone of Central Europe (Lithuania) provides results for the relevant optimization of agrarian landscapes in unproductive coarse-textured soils and helps to improve and protect the soil. This study focuses on the vegetation succession, pedogenesis, carbon stocks, and different soil nutrients of ex-arable Arenosols under renaturalization processes in Lithuania (East Baltic region), Central Europe. Research carried out on soil renaturalization in infertile soils can provide science-based recommendations for farmers working on low-productivity sandy soils. Considering the scope of activities and the anticipated production in the short or long term, farmers have the opportunity to choose the most appropriate land use that would also maintain stable soil fertility or even improve it.

5. Conclusions

This study reveals a shift in species composition between 1995 and 2022 in Arenosol, combined with declines in species richness and diversity, with the greatest rates of change occurring during the abandonment period. In the initial period of abandonment of the arable land, in the phytocenosis, plant species typical of phytocenoses of arable lands were partially preserved. After 6 years (in 2000), the species composition of herbaceous vegetation decreased by 36%. Comparing the years 2000 and 2004, the species composition of herbaceous vegetation increased by 24%. During the period 2004–2015, the species composition of herbaceous vegetation increased again by 46%, while during the period 2015–2022, it decreased by 32%. In this way, over the three decades of abandonment, the plant richness changed in an up-and-down manner.

As the duration of the abandonment increased, the number of perennial grasses species also increased, showing a rising curve over the 27-year period. In comparison with the first year of abandonment, perennial plant species had doubled by 2022.

After 10 years, perennial plant species began to dominate in the phytocenosis, mostly consisting of plants of the Poaceae family. During the research period, the species composition of the plants changed, and only plants of three families (Asteraceae, Poaceae, and Fabaceae) were constantly present in the phytocenosis. In total, over a 27-year period, 81 plant species grew on the abandoned site in different periods, contributing to the conservation of biological diversity in the surrounding area.

The productivity of the abandoned land’s phytocenosis increased as the duration of the abandonment period increased (r = 0.70). A positive relationship was established between the phytocenosis biomass and the number of cereal grass species (r = 0.44). The increase in productivity may also have been associated with the increase in temperature in the region, especially during the growing season. However, this hypothesis requires further research. No influence of the amount of precipitation on the yield of phytocenosis biomass was found (r = 0.001).

The succession of ex-agrogenic vegetation manifested in a directional change in the species composition of the vegetation, as well as in the formation of plant matter stocks. Plant matter stocks are shared by each plant species and, in this way, participate in the formation of phytomass and mortmass. In this study, new data were obtained to characterize the rate and intensity of the transformation of the chemical properties of ex-arable land in the Arenosol. The abandonment of land on low-productivity soil had a positive effect on soil fertility—the organic carbon concentration in the Ah horizon increased from 9.9 ± 0.08 to 14.5 ± 1.17 g kg⁻¹, the thickness of the Ah horizon increased by 3 cm, and the SOC stocks increased by 24.51 Mg ha⁻¹. Improving the chemical properties of the soil makes it possible to reintroduce this land for agricultural use for sustainable food production and economic returns for farmers.