Reducing drought vulnerability of forest soils using Xanthan gum-based soil conditioners

Abstract

Climate change increases the frequency and severity of droughts in many parts of Europe, thereby affecting the availability of water resources. Therefore, preserving the soil water content is essential for maintaining forest diversity and plant vitality. To improve soil hydraulic properties and reduce drought vulnerability, three xanthan-gum-based soil conditioners (SC_R, SC_CG, and SC_ZZC) were developed under the European ONEforest project. These soil conditioners (SCs), including oxide ash and cellulose fibres of different lengths, differ in their filler properties. This study evaluated the performance of these soil conditioners in forest soils in Slovenia (S1), Spain (S2), and Germany (S3). Water absorption, water retention, hydraulic conductivity, seed germination, and plant growth in the untreated soils and mixtures were analysed.

The results showed a 53 %–100 % increase in water absorption with high dosage of SCs (1.7 % of dry SC per dry soil mass). The addition of SCs also significantly improved the water retention capacity of the treated soils within a suction range of 0–100 kPa, which is advantageous for maintaining adequate water availability for plant growth. The effect of SCs on unsaturated hydraulic conductivity varies depending on the soil type. For example, at a similar suction, the unsaturated hydraulic conductivity of mixtures with soils S2 and S3 can be more than an order of magnitude lower than that of untreated soils S2 and S3. In contrast, mixtures with soil S1 showed unsaturated hydraulic conductivity similar to that of the untreated soil. In the seeding experiment, plants in treated soils survived for up to 8 d without watering compared to those in untreated soils, with survival linked to the initial water content of the mixtures. These findings suggest that soil conditioners reduce drought vulnerability by improving water retention and regulating water loss. However, the optimal dosage should be adjusted according to different soil types and local environmental conditions.

Highlights

• •Xanthan gum-based soil conditioners enhance the soil's water retention capacity.
• •Treated soils have lower hydraulic conductivity than untreated soils.
• •Xanthan gum-based soil conditioners extend plant survival in drought conditions.

1. Introduction

Trees and other forest plants play crucial roles in mitigating the effects of climate change on forest ecosystems. The absence of vegetation directly exposes forest soils to the adverse effects of rainfall events [1] and drying due to direct sunlight, thereby disrupting the natural water balance. Furthermore, the stability of a slope surface (erosion resistance), which often comprises well-drained and low-cohesion soil, depends on the mechanical reinforcement offered by trees and plant roots in the soil, as well as on the soil water retention characteristics [[2], [3], [4], [5], [6], [7]]. Ni et al. [8] and Ng et al. [9] developed advanced numerical models to quantify the effect of root hydromechanical reinforcement on slope stability.

Root distribution is influenced by the climatic regime and the ability of the soil to retain water, which is especially vital in regions where plant growth occurs under conditions with relatively limited water availability [3]. The application of biopolymers with water superabsorbent properties has shown promising results in terms of enhancing soil water retention capacity, thus mitigating the vulnerability of forest soils to drought [4,[10], [11], [12]]. Biopolymers not only improve the soil water retention capacity but also promote the germination and growth of vegetation [4,13]. Consequently, they contribute to creating a suitable and stable environment for plant and tree growth [4,10,[12], [13], [14], [15]].

Numerous studies have illustrated that the addition of biopolymers, with xanthan gum (XG) prevailing among them, alters the geotechnical properties of treated soil and effectively reduces the potential for soil erosion [[16], [17], [18], [19], [20], [21], [22], [23], [24], [25]]. Among the different biopolymers used for soil treatment, XG stands out because of its efficiency, availability in large quantities, and low price [26]. Moreover, the XG-treated soil can absorb large amounts of water during the rainy season and slowly release water during the dry season. Thus, XG treatment of soil plays a vital role in the preservation of soil water content and in improving vegetation growth and survival [27,28]. In addition to the beneficial effects of biopolymers on soil hydraulic properties, Zhang et al. [11] and Tariq et al. [13] emphasised the importance of optimising their dosage. Excessive use of biopolymers can lead to soil hardening at low water content and a reduction in soil porosity at high water content. Both can be unfavourable for seed germination and plant growth. The formation of a dry crust on treated soil can prevent water infiltration and evaporation, as well as the penetration of sprouts to the surface. Reduced soil porosity may restrict the availability of the oxygen necessary for root system development and growth. Therefore, when determining a suitable treatment procedure, the interactions between each polymer and the soil type should be considered. In a recent study [29] plant growth tests were performed using clayey soil treated with different XG dosages. The results revealed a significant peak in germination percentage and coverage of the plants in the soil treated with 2 % of XG (per dry soil mass). For cases with higher dosages of XG, the growth indicators were even lower than those of untreated soil, leading to the conclusion that 2 % of XG is the upper limit for clay treatment.

Although the saturated hydraulic properties of both treated and untreated soils have been extensively studied [11,26,30,31], unsaturated hydraulic properties have received far less attention [32], despite their greater importance under field conditions [33].

Al-Darby [34] investigated the impact of hydrophilic type of biopolymer Jalma gel at concentrations of 0.2 %, 0.4 %, and 0.8 % on the soil water retention curve (SWRC) of sandy soil. These findings suggest that the available water content increased exponentially and the saturated hydraulic conductivity decreased exponentially with increasing amount of Jalma gel. Previously, Mustafa et al. [35] determined that for a loamy sand soil, achieving a 50 % reduction in water penetrability is crucial to curb deep percolation losses while simultaneously maintaining an adequate infiltration rate through the soil. Consequently, Al-Darby's study recommended an optimal dosage of 0.4 % Jalma gel. This dosage successfully reduced deep percolation losses while maintaining satisfactory infiltration and water retention characteristics.

Tran et al. [4,36] used up to 1.0 % XG (to the dry mass of soil) to treat a sand-clay mixture. The experimental results show that the use of XG can improve the water-holding capacity of soil and control the rate of water loss from the soil under desaturation conditions.

Chang et al. [10] demonstrated that in the case of inorganic silty loam (ML), treatment with 0.5 % (to the mass of dry soil) of β-glucan and XG, proved to be a promising countermeasure for various aspects of desertification. These treatments were also effective in improving soil erosion resistance, retaining soil water content, and promoting cultivation. A notable improvement in the soil water retention capacity was also reported by Wang et al. [12] for low plasticity clay (CL) treated with 0.5 %–2.0 % per dry soil mass of XG, gellan gum, and guar gum and by Mohawesh and Durner [37] for sandy soil treated with 0.5 % of Luqua-sorb hydrogel.

Liao et al. [38] conducted a 120d experiment on sandy loam treated with different dosages of synthetic polyacrylamide and acrylic acid-based hydrogels. They found that these superabsorbent polymers substantially reduced the unsaturated hydraulic conductivities of treated soil samples, especially in the early stages.

Ng et al. [32] investigated the effects of biopolymers on the gas permeability of compacted clay with various densities and water contents. They found that the gas permeability of the biopolymer-treated compacted clay intended for landfill cover systems was consistently lower than that of untreated clay, regardless of the specimen density or water content. This reduction in gas permeability was attributed to the biopolymers, gellan and xanthan gum, which reduced larger soil pores and clogged smaller pores. Additionally, biopolymers absorb water and hydrate and further clog the soil pores. Such a decrease in gas permeability can disrupt the balance of air and water in the pores of forest soils, potentially leading to an imbalanced air-water cycle that may adversely affect plant growth and development.

Zhang et al. [31] highlighted the importance of understanding changes in the SWRC as a biopolymer-treated soil transitions from a saturated to an unsaturated state. Drying alters the water retention capacity of biopolymer-treated soils. Hydrogels, which can retain bound water for extended periods, gradually release water under extremely dry conditions, thereby increasing the cross-sectional area for water movement. Because field soils to which hydrogels are to be applied are often water-limited and remain in an unsaturated state, further studies must be conducted on the impact of hydrogel applications on unsaturated hydraulic conductivity [33].

This study primarily aimed to conduct a comprehensive assessment of the effect of newly developed XG-based soil conditioners (SCs) on the hydraulic properties of various forest soils, with a particular focus on the behaviour of SC-treated soils under unsaturated conditions. The uniqueness of the XG-based SCs developed within the ONEforest project frame is the enhanced XG matrix with oxide ash and cellulose fillers of various fibre lengths. Therefore, we focused on identifying how different fillers influence the effectiveness of SCs in soil treatment.

After the development of SCs [39], investigation of their rheological properties, initial basic application in soil conditioning [15] and influence of SCs on the geotechnical properties (drained shear strength, compressibility and hydraulic conductivity) of treated soils in a saturated state [40], the present study represents a follow-up research focused on the efficacy of SCs in enhancing soil hydraulic characteristics in saturated and unsaturated states.

To illustrate the efficacy of SCs in enhancing soil hydraulic characteristics in saturated and unsaturated states, three different forest soil types sourced from various European locations were utilised. The water retention characteristics of both the untreated and treated soils were analysed across a wide suction range using Hyprop and WP4-T devices (METER group). The saturated hydraulic conductivity was measured using a falling head permeameter (oedometer), whereas the unsaturated hydraulic conductivity was estimated from Hyprop measurements. Additionally, the influence of SCs on the seed germination and plant growth was investigated under laboratory conditions.

2. Materials

2.1. Soils

Laboratory investigations were conducted on three types of soil from different forest areas in Europe: Ljubelj in the Alpine part of Slovenia (S1), Catalonia, northeastern Spain (S2), and Heldburg, Germany (S3). By utilising these diverse forest soils, the influence of inherent soil properties on the interactions of soils with different SCs treatments was assessed.

Although the samples were collected from the soil layer beneath the fresh organic layer on the surface, they still contained a small amount of older and partially decomposed organic matter. The organic content of soil S1 was 8.2 % by mass per dry matter (DM), whereas S2 and S3 had organic contents ranging from 1.8 % to 5.2 % DM.

Prior to the laboratory investigations, thesoil samples were manually cleared of small tree branches, needles, and leaves. Fig. 1 illustrates the cleared soil samples, and Table 1 summarises their index properties. A more comprehensive description of the geotechnical properties of soils, including laboratory test methods, can be found in Ref. [40].

Fig. 1.

Investigated soils after removing older and partially decomposed organic particles.

Table 1.

Index properties of the different soil types.

Soil Type		S1	S2	S3
Natural water content, w (%)		43	10	18
Particle density,${\mathrm{\rho }}_{\mathbf{s}}$(g/cm3)		2.52	2.59	2.56
Liquid limit, wL(%)		67	31	26
Plastic limit, wP(%)		41	15	13
Particle size distribution (%)	> 2.0 mm	3.81	3.17	1.20
	0.063–2.0 mm	13.12	39.03	36.38
	< 0.063 mm	83.07	57.80	62.42

2.2. Soil conditioners

Three types of XG-based SCs with various types of fillers were investigated. The SCs were selected based on their performance during the development phase and initial application testing [15,39]. Table 2 summarises the properties of the fillers used in the preparation of SCs. The fillers were mixed with 4 % XG-water solution at the dosage of 2 %. The mixing operations were performed using a Dispermat® F1 mixer (VMA-Getzmann Gmbh, Reichshof, Germany), operating at 5000 rpm for 15 min, until a homogeneous mixture without lumps was obtained. After mixing, the SCs were dried in an oven at 50 °C for 72 h and ground in a Piovan® RN166/1 granulator (Piovan SpA, Venice, Italy) for 3 min. Detailed descriptions of the development and properties of SCs can be found in Refs. [15,39].

Table 2.

Composition of different fillers in soil conditioners [15,40].

SC label	Filler type	Cellulose content (%)	Oxide ash content (%)	Average fibre length (μm)	Aspect ratio
SC_R	Arbocell R	>99	0.5	200–300	9.9
SC_CG	Cellugrün	80	15	1400	31.1
SC_ZZC	Arbocell ZZC 500	80	15	400	8.8

Fig. 2 depicts the SCs in three representative states: dry product (a), hydrogel after saturation (b), and thin crust of SCs after drying at room temperature (c) where it can be seen that after drying the hydrogel forms very stiff and fragile dry crust.

Fig. 2.

The soil conditioners in (a) dry state, (b) saturated state - hydrogel and (c) dry crust resulting from the air drying of the hydrogel.

2.3. Soil treatment procedure

The forest soils were treated with both low (L) and high (H) doses of the soil conditioner SC_R, whereas SC_CG, and SC_ZZC were applied solely at high doses. A high dosage of SC corresponded to 1.7 % of dry SC per dry soil mass, whereas a low dosage corresponded to 0.4 % of dry SC per dry soil mass. The four mixtures listed in Table 3 were prepared and investigated for each soil type. Therefore, 12 mixtures were included in this study.

Table 3.

Investigated mixtures [40].

Soil	Soil Conditioners	Dosage	Mixture Label
S∗	SC_R	low, L, 0.4 %	S∗+SC_R L
	SC_R	high, H, 1.7 %	S∗+SC_R H
	SC_CG	high, H, 1.7 %	S∗+SC_CG H
	SC_ZZC	high, H, 1.7 %	S∗+SC_ZZC H

S∗—represents the identification number of the soil (S1, S2, and S3).

Mixtures of selected soil and SCs for the determination of water absorption w_A were prepared using dry soil (T_drying = 45 °C) and dry SC. Prior to mixing, both the dry soil and SC were ground to achieve a maximum particle diameter of 0.355 mm. For other laboratory investigations, soils were used in their natural state, preserving their natural water content and grain size distribution (Table 1), and mixed with dry SCs. Each specimen was prepared from a fresh mixture.

3. Methods

Laboratory investigations of untreated soils and mixtures were conducted following standard procedures for determining soil properties or by employing methods recommended by the manufacturers of the devices used. Owing to the unique properties of mixtures influenced by soil conditioners SCs, certain adjustments were made to the preparation of the specimens and test methods. As mandated by the standards, detailed descriptions of the deviations from the test methods are provided below.

3.1. Gravimetric water content

The gravimetric water content was determined by drying soils and mixtures to a constant mass in a ventilated drying oven maintained at 45 °C. The drying temperature was selected based on the proposal of the standard for soils containing organic matter [41].

3.2. Water absorption

Water absorption tests were performed on untreated soils and mixtures using an Enslin–Neff device. The standard [42] recommends a dry specimen mass of 1.0 g for specimens with water absorption lower than 100 %, whereas for specimens with water absorption equal or higher than 100 %, the suggested initial mass of dry specimen is 0.2 g. Due to the difficulty in preparing a homogeneous representative mixture with a mass of 0.2 g, tests were conducted on specimens with a dry mass of 1.0 g even in cases where the water absorption exceeded 100 %. Following the recommendations of the standard, water absorption was assessed after 24 h (w_{A 24h}). At least two test replicates were conducted for each untreated soil sample and mixture. To assess the effect of the initial mass of the dry specimens on the test results, tests for randomly selected mixtures (S1+SC_R H, S2+SC_ZZC H and S3+SC_CG H) were repeated using specimens with an initial dry mass of 0.2 g.

Because of extremely high water absorption of SCs alone, water absorption tests using the Enslin-Neff device could not be conducted. Consequently, an adapted method was used to determine the water absorption of the SCs, as shown in Fig. 3. A dry filter paper of known mass was positioned inside a Petri dish of known mass and connected to a water basin using a filter paper strip. After the saturation of the filter paper, the masses of the saturated filter paper and Petri dish was determined. Approximately, 1.0 g of the dry specimen of SCs, soil, or mixture was carefully placed on a saturated filter paper. The water-absorption capacities of the specimens were determined by weighing them at various intervals during the absorption process. The test was terminated if no discernible difference was found in the mass between two consecutive weightings, indicating that the absorption process had reached a stable state. Additionally, the water content of each specimen was measured at the end of the test.

Fig. 3.

Schematic representation of the set up for adapted water absorption test.

To validate the test results, the adapted method was also used for water absorption tests of the untreated soil S2 and its mixtures.

3.3. Soil water retention curve (SWRC)

Soil water retention curves (SWRCs) were determined using the Hyprop evaporation method device (for matric suctions up to approximately 100 kPa) and a dew point potentiometer WP4-T (for total suctions exceeding approximately 300 kPa). Investigations were conducted following the manufacturer's instructions [43,44] and adhering to standard procedures [45].

The Hyprop device involves measurements of matric suction at two depths (1.25 cm and 3.75 cm) within a specimen with the diameter of 8 cm and height of 5 cm. The time-dependent evaporation rate was obtained by the automated weighing of the initially saturated specimen. The changes in weight were assumed to be equal to changes in water mass. This means that biodegradation during the investigation, lasting for at least one month, was not considered. The water retention values were estimated from the average gravimetric water content and average pressure measured during drying of the specimen using two tensiometers.

Specimens were prepared from untreated and treated soils (mixtures). Homogeneous mixtures were prepared from soil at natural water content with the addition of dry SC and were slightly compacted into the Hyprop mould. The specimens in the mould were then saturated by immersion in water up to four-fifths of their height. Volume changes were minimised by placing porous discs on the upper and lower surfaces of the specimens and applying a weight of 10 kPa to the top during the saturation process. After saturation was completed, tensiometers mounted on the Hyprop device were pushed into the specimen and measurements were started. To provide quasi-steady-state conditions (constant flux and hydraulic gradient over the time interval), the evaporation rate was reduced by covering the specimens with a perforated plastic bag of a known mass.

In the high suction range, the SWRC was measured using a potentiometer WP4-T. The potentiometer WP4-T was calibrated each day before regular measurements using standard solution of 0.5 M KCl. Measurements were conducted on the remaining portion of the sample (soil or mixture) prepared for the specimens investigated using the Hyprop device. Using drippers, the sample was wetted to a water content at or slightly below full saturation. The first specimen was prepared and the total suction was measured. The samples were then slowly dried stepwise at room temperature (22°) and the specimens were taken at various drying intervals. Prior to conducting the measurements, the prepared specimens were tightly sealed in special plastic containers for at least 24 h to ensure that the water content and suction were equilibrated.

3.4. Saturated hydraulic conductivity

The saturated hydraulic conductivity (k_sat) was measured using a falling-head permeameter [46]. These tests were conducted at the end of the 25 kPa loading stage as part of the incremental loading oedometer tests [40].

3.5. Unsaturated hydraulic conductivity

The unsaturated hydraulic conductivity (k_unsat) of the soils and mixtures was derived from the data measured using a Hyprop device. Unsaturated hydraulic conductivity reflects the ability of the soil to transport water from the wet bottom to the top of the specimen, where water evaporates. Based on the loss of mass per time interval and the difference in pressures measured with tensiometers, the water flux and unsaturated hydraulic conductivity were determined according to the simplified evaporation method presented in Refs. [[47], [48], [49]]. The calculations were performed using LABROS Soilview and Soilview-Analysis Software [50] and were based on the following assumptions: (1) changes in hydraulic conductivity within the specimen during the measurement time interval were negligible; (2) the flux at a depth of 2.5 cm within the specimen was half of the total flux and could therefore be calculated from the total evaporative soil water volume difference in the time interval; and (3) the vertical distributions of water content and suction were linear throughout the entire column.

$${\mathrm{k}}_{\text{unsat}}\left(\stackrel{═}{s_i}\right)=\frac{q_i/2}{1-\raisebox{1ex}{${\Delta s}_i$}\!\left/ \!\raisebox{-1ex}{$\left({\gamma }_w\Delta z\right)$}\right.}\text{,}$$ (1)

where $\stackrel{═}{s_i}$ is the mean suction (in kPa) of the two consecutive time steps (i, i-1) at two depths, defined by positions of the tensiometers within the specimen $z_1$ and $z_2$ ($\stackrel{═}{s_i}=\left(s_{1,i}+s_{2,i}+s_{1,i-1}+s_{2,i-1}\right)/4$). The ${\Delta s}_i$ is the difference between the tensiometer readings (${\Delta s}_i=\left(s_{2,i}-s_{1,i}+s_{2,i-1}-s_{1,i-1}\right)/2$), ${\gamma }_w$ unit weight of water and $\Delta z$ is determined as ${\Delta z=z}_2-z_1$. The water flux is denoted as $q_i$ ($q_i=\left(m_i-m_{i-1}\right)/\left(\left(t_i-t_{i-1}\right){\rho }_{w}A\right)$, where $m_i$ is the mass of specimen at time $t_i$, $A$ is cross-section area of the specimen, and ${\rho }_w$ is water density).

3.6. Plant growth

Seeding experiments were conducted to assess the effects of SCs on seed germination and plant growth. Christmas wheat (Fig. 4, left) was selected for the seeding experiment because of the expected fast germination and growth rates. The investigation was carried out in laboratory conditions with temperature of 19–24 °C and a relative air humidity between 40 and 60 %. Specimens of untreated soils with natural water content and mixtures with high dosages of different SCs were placed in plastic containers. Seventy-eight seeds were evenly distributed across the surface of the soil or mixture within each container (Fig. 4, right) and covered with approximately 1 cm thick layer of soil or mixture. The specimens were subsequently watered with tap water. Two parallel experiments were performed:

• (1)Specimens with the same initial water content: Each untreated soil specimen was watered until it could no longer absorb water. The amount of added water was determined by weighing the specimens. The mixtures were watered with the same amount of water as the untreated soil.
• (2)Saturated specimens: Each specimen was watered until it could no longer absorb water. Excess water either exited the bottom or was stagnant on the surface (referred to as saturated specimens).

Fig. 4.

Christmas wheat seeds (left) and seeds placed in mixture of soil S2+SC_R H before covering the seeds with approx. 1 cm thick layer of mixture (right).

During the experiment, seed germination and plant growth were monitored using photographs. The gravimetric water content was calculated from the measured mass of the specimens at different time intervals and the final dry mass. Saturated specimens were watered once more after 7 d, whereas specimens prepared with the same initial water content were watered upon initial germination and then twice more.

4. Results

4.1. Water absorption

Fig. 5 shows the water absorption capacity of the SCs granules determined using the adapted method illustrated in Fig. 3. For an improved overview of the results and easier traceability, the point measurements are connected by curves. Water absorption rates were virtually equal for all three types of SCs. The final water absorption was achieved after 145 h and exceeded 1500 %. Measured values fall within the same range as those reported by Sorze et al. [39], who conducted investigations on dry specimens exposed to constant humidity (RH = 80 %) and temperature (25 °C) for 100 h.

Fig. 5.

Determination of water absorption capacities of the three types of SCs using the adapted method.

The water absorptions of the soils and mixtures with an initial dry mass of 1.0 g per specimen are summarised in Table 4. Further details are provided in Ref. [40]. At least two test repetitions were conducted for each soil or mixture, and the average of the measurements after 24 h was considered the water absorption (w_{A 24 h}).

Table 4.

Water absorption of specimens from soils and mixtures with the initial dry mass of 1.0 g (0.2 g). ∗representing the identificatory number of the respective soil.

Soil	S1	S2	S3
Specimen	wA24 h (%)
S∗	88	68	50
S∗+SC_R L	95	73	62
S∗+SC_R H	158 (126)	105	91
S∗+SC_CG H	145	103	100 (112)
S∗+SC_ZZC H	135	105 (93)	101

For mixtures with a high dosage of SCs, w_{A 24 h} was generally close to or higher than 100 % (Table 4). Consequently, tests on mixtures S1+SC_R H, S2+SC_ZZC H and S3+SC_CG H were conducted also on specimens with an initial dry mass of 0.2 g, as recommended by the standard. Regardless of the initial dry mass of the specimens, the measured water absorption was within the same range (Table 4; values in parentheses). This confirmed that the water absorption capacity obtained with an initial dry mass of the specimen 1.0 g was reliable for the investigated specimens.

The efficiency of SCs depends on their dosage and the soil type. A low dosage of SCs resulted in 7–8 % relative increase in w_{A 24 h} compared to the w_{A 24 h} of untreated soils S1 and S2, whereas for S3, the increase was 24 %. A high dosage of SCs was least effective in the case of soil S2, where 51–54 % increase in w_A was observed regardless of the type of SC. On the contrary, a high dose of SCs was most effective in the case of soil S3, where the w_{A 24 h} of mixtures ranged from 82 % (SC_R) to 102 % (SC_ZZC) higher than w_{A 24 h} of the untreated soil.

Fig. 6 depicts a comparison of the results obtained through both the standard procedure (solid bars) and the adapted test method (hatched bars) for untreated soil S2 and its mixtures with SCs. The tests were conducted on specimens with an initial dry mass of 1.0 g. The adapted test method showed reduced water absorption for untreated soil compared with the standard Enslin-Neff test method. However, for the mixtures, the results obtained using the two methods were comparable.

Fig. 6.

Comparison of water absorption capacities of untreated soil S2 and its mixtures determined using the Enslin-Neff test method and the adapted method (1.0 g of the specimen).

4.2. SWRC

The Soil Water Retention Curves (SWRCs) for untreated soils and mixtures are depicted in Fig. 7. The data obtained from the Hyprop device are represented by a solid line, whereas the WP4-T device results are illustrated by scatter plots.

Fig. 7.

SWRCs for untreated soils S1 to S3 and their mixtures.

Specimens from the mixtures and the corresponding untreated soils were prepared with comparable initial dry densities. Table 5 presents the initial dry densities and the dry densities of the specimens from the untreated soils and their mixtures after saturation and swelling. Owing to swelling during saturation, the specimens from the mixtures exhibited lower dry densities and higher water contents than those from the untreated soils at the start of the test. Similar behavior is also expected under field conditions, because the vertical stress will generally be lower than that applied during the saturation of specimens in the laboratory (10 kPa). Thus, the measured SWRCs are valid for describing the effect of SCs on water retention capacity.

Table 5.

Initial and post-saturation dry densities of specimens from untreated soils and their mixtures. ∗representing the identificatory number of the respective soil.

Soil	S1	S2	S3
Specimen	ρd initial/ρd post-saturation (Mg/m3)
S∗	1.04/1.04	1.29/1.29	1.20/1.20
S∗+SC_R L	1.04/0.94	not investigated	not investigated
S∗+SC_R H	1.04/0.95	1.29/1.14	1.22/1.14
S∗+SC_CG H	1.05/0.96	1.28/1.11	1.21/1.18
S∗+SC_ZZC H	1.05/0.91	1.29/1.14	1.22/1.18

The lowest measured suctions of the soil and its mixtures varied, depending on the initial saturation of the specimens. For both soils S1 and S2 and their mixtures, the achieved initial saturation of specimens was higher than 90 % (often exceeding 95 %). However, for specimens from mixtures with S3, achieving an initial saturation higher than 70 % (S3 + SC_R H) and 85 % (S3 + SC_ZZC H) was unattainable, even after exposing the specimens to water for up to 7 d. The saturation time was limited to 7 d owing to the presence of organic matter in the mixtures. The prolonged saturation time, and in the case of mixtures with S3, the lower initial saturation achieved, can be attributed to the XG contained in the SCs. Upon contact with water, XG forms a viscous hydrogel that coats soil particles and blocks pores within the soil matrix, as reported by Wang et al. [12]. This led to a significant reduction in water flow through the treated soil specimens.

Fig. 7 shows that the drying curves of specimens from mixtures in the low-suction range (<100 kPa) consistently occupied higher positions compared to the SWRCs of untreated soils. This consistent upward shift implied an enhancement in the water retention characteristics of the treated specimens within a particular suction range. In addition, the results obtained for the S1 mixture with two different dosages of SC_R indicated that the water retention capacity of the mixtures increased with increasing SC content. Some improvement in the water retention capacity was also observed in the middle suction range (100 kPa–2000 kPa). In this suction range, the distinctions between SWRCs of soil and mixtures are minimal, potentially stemming from various factors such as the presence of SCs, inhomogeneity of specimens, or the declared accuracy of the WP4-T device, specified at ± 100 kPa for suctions below 1000 kPa. When the suction exceeded 2000 kPa, which was higher than the permanent wilting point [51], the water retention capacity of the mixtures showed no noticeable difference compared with the untreated soils.

4.3. Hydraulic conductivity

Fig. 8 presents both the saturated hydraulic conductivity measured using a falling head permeameter (solid rectangles) and the unsaturated hydraulic conductivity calculated from the Hyprop measurements (empty circles) of untreated soils and mixtures. Irrespective of the soil type and saturated hydraulic conductivity of the untreated soils, the introduction of a high dosage of SCs resulted in saturated hydraulic conductivity values ranging from 10⁻⁹ m/s to 10⁻¹⁰ m/s. This implies a reduction in the saturated hydraulic conductivity by two to four orders of magnitude compared to that of untreated soils.

Fig. 8.

Saturated and unsaturated hydraulic conductivity for untreated soils S1, S2 and S3, and mixtures with high dosages of SCs.

Although the influence of biopolymers on the saturated hydraulic conductivity has been well investigated, data on the unsaturated hydraulic conductivity of forest soils treated with biopolymers are lacking. Unsaturated hydraulic conductivity is primarily influenced by the water content or matric suction of the soil. In addition, it is contingent on the structure and connectivity of the pore spaces within the soil matrix. As illustrated in Fig. 8, the unsaturated hydraulic conductivity generally decreased with increasing matric suction and was generally lower than the saturated hydraulic conductivity. Notably, for the untreated soils, this decrease was more pronounced than that for the mixtures. Untreated soil S1 and its mixtures exhibited comparable k_unsat values for higher matric suction, whereas untreated soils S2 and S3 showed higher values of k_unsat than their mixtures for the same matric suction. There were no notable differences between the mixtures of different types of SCs with soil S1, whereas in the case of the soils S2 and S3, the influence of the type of SC was visible. The results indicated that SCs affected both the saturated and the unsaturated hydraulic conductivity in various types of soils, with no discernible influence attributable to the specific type of SCs used.

4.4. Seed germination, plant growth and vegetation viability under drought conditions

Fig. 9, Fig. 10, Fig. 11 illustrate the average gravimetric water content measured in the specimens prepared from various soils and mixtures with a high dosage of SCs during the investigation of seed germination, plant growth, and wilting in the drying phase. For each mixture, two separate graphs were plotted: (1) for specimens with the same initial water content corresponding to the water retention capacity of untreated soil, and (2) for initially saturated specimens watered until the soil or mixture could not absorb water. The germination and wilting points are marked with circles and diamonds, respectively. Germination and wilting points were determined based on the first seeds to germinate and the first plants to wilt. Photographs of the initial states of the specimens (during 1st watering), seed germination, and wilting are included in the Supplementary File (Appendix A).

Fig. 9.

Gravimetric water content of soil S1 and mixtures with high dosage of SCs during seed germination (circles), plant growth and wilting during drying (diamonds): (1) specimens with the same initial water content and (2) initially saturated specimens.

Fig. 10.

Gravimetric water content of soil S2 and mixtures with high dosage of SCs during seed germination (circles), plant growth and wilting during drying (diamonds): (1) specimens with the same initial water content and (2) initially saturated specimens.

Fig. 11.

Gravimetric water content of soil S3 and mixtures with high dosage of SCs during seed germination (circles), plant growth and wilting during drying (diamonds): (1) specimens with the same initial water content and (2) initially saturated specimens.

Regardless of the soil type and mixture, germination in specimens with the same initial water content began within 3 or 4 d of seeding. However, for the saturated mixtures S1+SC_R H, S3+SC_CG H, and S3+SC_ZZC H, the germination time was extended from 5 to 8 d. In contrast, for the other saturated specimens, the germination time remained within 3–4 d. Upon watering, a thin layer of hydrogel formed on the surfaces of the saturated specimens from mixtures. However, upon air-drying, the mixture on the surface transformed into a rigid crust, impeding the penetration of sprouts during the initial stages of growth. Watering at distinct time intervals, represented by the vertical dashed blue lines in Fig. 9, Fig. 10, Fig. 11, resulted in the crust becoming wet and soft.

The germination rate and plant growth depend on the type of soil and, in the case of mixtures, on the type of SCs and the initial water content. Fig. 12 illustrates that, under similar laboratory conditions and care, untreated soil S3 exhibited the lowest germination rate and plant growth, whereas untreated soils S1 and S2 demonstrated comparable results.

Fig. 12.

Germination and growth of Christmas wheat in different types of untreated soils with denoted time after seeding.

Fig. 13 illustrates the plant growth and germination rates in the untreated soil S3 and its mixtures prepared with the same initial water content (1) and in a saturated initial state (2), 187 h after seeding. Specimens with the same initial water content (1) showed a favourable effect of SCs, resulting in higher plant growth than in an untreated soil. However, excluding the mixture with SC_R H, the influence of SCs in the initially saturated specimens was less favourable, as evidenced by reduced seed germination and consequently poorer plant growth. Whereas the initially saturated mixture S3+SC_ZZC H exhibited fewer promising results, the saturated mixtures of soils S1 and S2 with the same soil conditioner (Fig. 14) showed more favourable outcomes. Fig. 15 shows plants grown in mixtures prepared from soils S2 and S3 with SC_R and SC_CG at 267 h post-seeding and watering to the same initial water content (1). Plants in the soil S3 mixtures display diminished height compared with those in the S2 mixtures, irrespective of the SC type.

Fig. 13.

Influence of the SCs on the growth of Christmas wheat on example of soil S3 for specimens with the (1) same initial water content and (2) initially saturated specimens. Photos were taken 187 h after seeding.

Fig. 14.

Growth of Christmas wheat in mixtures of soils S1, S2 and S3 with high dosage of SC_ZZC for specimen with the (1) same initial water content and (2) initially saturated specimens with denoted time after seeding. Pictures were taken at the peak of the plant growth.

Fig. 15.

Height of Christmas wheat in the mixtures of (a) S2+SC_R H, (b) S2+SC_CG H, (c) S3+SC_R H and (d) S3+SC_CG H at the peak of the plant growth. Specimens from the same soil were prepared at the same initial water content.

During the non-watering phase, plants that germinated in the treated soils survived longer than those that germinated in the untreated soils. The specimens that were saturated at the beginning of the test and had the largest water supply showed the best plant survival (Fig. 9, Fig. 10, Fig. 11, diamonds). Interestingly, the wilting point was identified at a gravimetric water content corresponding to a suction of approximately 1500 kPa, which has been reported as the permanent wilting point [51]. Water content at suction 1500 kPa was approximately 20–23 % for S1 and its mixtures, 11 % for S2 and its mixtures, and 6–8% for S3 and its mixtures (Fig. 7). Fig. 9, Fig. 10, Fig. 11 (black horizontal dashed line) show that these water content values correspond well with the empirically assessed wilting points during the experiment.

5. Discussion

In this study, we investigated the use of xanthan-gum-based SCs to improve drought resilience in forest soils. The results showed clear improvements in water absorption and water retention capacities in the SC-treated soils, which are crucial for maintaining soil moisture and supporting plant growth during drought. Water absorption capacity increased by 53 %–100 % depending on soil type and SC dosage, while the most substantial enhancements in water retention occurred at suctions below 100 kPa. Although the investigated soils were fine-grained, their liquid and plastic limits and grain size distributions varied. Soil S1, which was classified as silt with sand, had the highest liquid and plastic limits, water absorption, and fines content. Consequently, the water absorption and retention capacities of mixtures containing S1 were the highest. By examining these diverse forest soils, we demonstrated that the inherent properties of each soil type significantly influenced the mixture characteristics. Given the lack of prior research on the application of biopolymers to forest soils, exploring multiple soil types is essential to gain a comprehensive understanding of the effects of SCs.

These findings are consistent with previous research indicating that biopolymers, such as xanthan gum, enhance soil water absorption and retention [4,10,11,28,29,31,[52], [53], [54]]. Regardless of the soil type and the properties of the additives used, the soil conditioners were the most effective in the lower suction range (<100 kPa). Some improvement in water retention capacity was also observed in the suction range between 100 kPa and 2000 kPa, although the effect was less pronounced [53]. Enhanced water retention in treated soils increased the availability of water to plant roots under drought conditions.

Moreover, this study highlights the impact of SCs on hydraulic conductivity. The treated soils exhibited lower saturated hydraulic conductivity than the untreated soils, which is consistent with previous studies [27,55,56]. In forests and other natural settings, the achievement and maintenance of soil saturation is infrequent and typically has a short duration. Therefore, the unsaturated hydraulic conductivity of the topsoil is a crucial parameter for studying water flow in the vadose zone [[57], [58], [59]]. In addition to providing water for plants and trees, the unsaturated zone also provides oxygen and nutrients to the plant roots [60]. Determining the unsaturated hydraulic conductivity is challenging, and measured data are scarce [61]. Although a direct comparison with literature data is not possible, the results indicate that SCs affect unsaturated hydraulic conductivity, with no discernible influence attributable to the specific type of SC used. For soils S2 and S3, the unsaturated hydraulic conductivity of mixtures was lower than that of the untreated soils, suggesting that SCs can effectively reduce water loss through evaporation and deep percolation.

The benefits of SCs also extend to plant growth under drought conditions, as demonstrated by the survival of plants in treated soils lasting up to eight days longer during non-watering phases. This observation is supported by Tran et al. [4] and Tariq et al. [13], who demonstrated that biopolymers can improve soil moisture availability and promote plant growth. However, upon air drying, the mixture on the surface can transform into a rigid crust, impeding sprout penetration during the initial stages of growth. Wang et al. [12] reported the formation of a dry crust on the surface of XG-treated soil and its effect on sprout penetration during the early growth stages.

The observed behaviour of the mixtures stems primarily from the hydrogel nature of xanthan gum, which forms upon contact with water. Thus, the influence of the filler property in the SCs on the properties of the mixtures was not recognised.

Despite these advantages, careful consideration is necessary when determining the dosages of SC under specific conditions [11,13]. The dosage cannot be accurately predicted based solely on intrinsic soil properties such as grain size distribution, liquid and plastic limits, or water absorption. Local climatic conditions, including precipitation and evaporation, as well as the intended purpose of use must also be factored into dosage decisions.

The SCs developed under the ONEforest project are in granular form, making them easy to apply in practice. Based on laboratory-scale production methods [15,39], the estimated cost is approximately 20–25 €/kg. This is comparable to the prices of similar commercially available products, suggesting the potential for scalable technology [40]. Similar to commercial products, the developed dry SCs can be mixed directly into the soil at the tree-planting site, focusing on the root zone, where moisture retention and nutrient uptake are critical.

Future studies should focus on long-term field studies to validate these findings and explore how SCs interact with various soil types and microbial communities to provide deeper insight into their mechanisms of action [62].

6. Conclusions

Laboratory tests on three untreated soils (S1, S2, and S3) and their mixtures with xanthan-gum-based SCs (SC_R, SC_CG, and SC_ZZC) yielded the following key conclusions:

• (1)Soils treated with SCs exhibited higher water absorption than untreated soils. While dosage and soil type influenced absorption, SC type had no effect.
• (2)When the suction was less than 100 kPa, the water retention capacity of the SC-treated soil was notably higher than that of the untreated soil. SC type was not identified as an influential factor. The improvement in water retention capacity is favourable for providing available water for plant growth.
• (3)SCs reduced the saturated hydraulic conductivity in all soils by two to four orders of magnitude, owing to pore clogging from swelling. Under unsaturated conditions, the impact of SCs varies. In soil S1, the effect is smaller than that in soils S2 and S3, where SC-treated soils have lower unsaturated conductivity.
• (4)During the non-watering phase, plants grown from seeds sown in treated soils generally had longer survival rates than those grown in untreated soils, with the best survival observed in specimens that were initially saturated and had the highest water supply.

CRediT authorship contribution statement

Jasna Smolar: Writing – original draft, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization. Barbara Fortuna: Writing – original draft, Visualization, Investigation, Data curation. Janko Logar: Writing – review & editing, Funding acquisition. Alessandro Sorze: Writing – review & editing, Resources, Methodology, Investigation. Francesco Valentini: Writing – review & editing, Resources, Methodology, Investigation. Matej Maček: Writing – review & editing, Investigation, Formal analysis, Data curation. Boštjan Pulko: Methodology, Resources, Supervision, Writing – review & editing.

Data availability statement

Data are available upon reasonable request to the corresponding author.

Funding

The study has been funded by the European Union's Horizon 2020 Research and Innovation Program within the project ONEForest: A Multi-Criteria Decision Support System for Common Forest Management to Strengthen Forest Resilience, Harmonize Stakeholder Interests, and Ensure Sustainable Wood Flows (Grant Agreement No.: 101000406) and by the research program P2-0180 (B) Water Science and Technology, and Geotechnical Engineering: Tools and Methods for Process Analyses and Simulations, and Development of Technologies, supported by the Slovenian Research Agency.

Declaration of competing interest

The authors have no conflicts of interest to declare that are relevant to the content of this article.

Appendix A. Supplementary data

The following is the Supplementary data to this article: