Soil & Tillage Research 233 (2023) 105811 Contents lists available at ScienceDirect Soil & Tillage Research journal homepage: www.elsevier.com/locate/still Quantifying root-induced soil strength, measured as soil penetration resistance, from different crop plants and soil types Francis Kumi a, Peter B. Obour b, Emmanuel Arthur c, Stephen E. Moore d, Paul A. Asare e, Joel Asiedu e, Donatus B. Angnuureng f, Kofi Atiah g, Kwadwo K. Amoah e, Shadrack K. Amponsah h, Selorm Y. Dorvlo i, Samuel Banafo e, Michael O. Adu e,* a Department of Agricultural Engineering, School of Agriculture, College of Agriculture and Natural Sciences, University of Cape Coast, Cape Coast, Ghana b Department of Geography and Resource Development, University of Ghana, Accra, Ghana c Department of Agroecology, Faculty of Technical Sciences, Aarhus University, Blichers Allé 20, Tjele DK-8830, Denmark d Department of Mathematics, School of Physical Sciences, College of Agriculture and Natural Sciences, University of Cape Coast, Cape Coast, Ghana e Department of Crop Science, School of Agriculture, College of Agriculture and Natural Sciences, University of Cape Coast, Cape Coast, Ghana f Africa Centre of Excellence in Coastal Resilience, Centre for Coastal Management, University of Cape Coast, P.O. Box UC56, Cape Coast, Ghana g Department of Soil Science, School of Agriculture, College of Agriculture and Natural Sciences, University of Cape Coast, Cape Coast, Ghana h CSIR-Crops Research Institute, Kumasi, Ghana i Department of Agricultural Engineering, University of Ghana, Accra, Ghana A R T I C L E I N F O A B S T R A C T Keywords: A common soil mechanical property for assessing soil strength is soil penetration resistance (PR) or soil cone Cone penetration index index (CI), which is related to the undrained shear strength of saturated and cohesive soil. Plant roots can in- Napier grass crease soil strength, but physical conditions may confound this. Pot experiments were conducted using 70 cm soil Nature-based bioengineering columns, three soil types (beach sand, erosion-prone soil, and two typical arable soils), and four crop plants Root system architecture Soil stabilization (maize, sorghum, Napier, and vetiver grass). We tested the hypothesis that plant roots impact soil strength, Vetiver grass measured as soil PR, and the induced soil strength differs based on plant species. The CI and root system ar- chitecture (RSA) traits were measured. Napier grass grown in arable soils recorded higher total biomass. Together with maize, Napier grass had a more significant root length density, particularly at 25–40 cm depth. The CI increased with increasing depth, with a 57–99% increase in CI in the bottom layer compared to the top layer of the soil column. The overall CI of soils grown to Napier grass (2.0 and 2.3 MPa) and maize (1.7 and 2.2 MPa) were similar, but both were higher than the soils cultivated with the other crop plants and unplanted control. The overall CI of the SEA sand of ~2.0 MPa was 36%, higher than that for the arable soils. Soil moisture content did not significantly increase CI, but the interaction of soil bulk density and root system traits could be implicated in increased CI of root-permeated soils. It is concluded that (i) roots growing in arable soils can in- crease CI and hence soil strength, possibly due to the binding effect of root systems, even when the transpiration effect of plants on soil moisture is low; (ii) crop plants contribute differently to soil strength, and (iii) Napier grass could offer a rapid growth and establishment option when considering plants for soil reinforcement and stability. 1. Introduction involves aspects of various components of soil strength. Soil shear strength (τ) describes the resistance to shear stress of a given soil, the Soil penetration resistance measures the soil’s resistance to defor- maximum stress soil can bear before failure. Typically, τ is a function of mation or compaction. It is the force required to penetrate the soil sur- three major parameters (τ = c + σtanφ), including the normal stress face or a specific depth within the soil. It is due to the cohesive forces acting on the failure surface (σ; kPa), the cohesion (C; kPa), and the exerted between individual soil particles and the frictional resistance angle of internal friction (φ; ◦) (Forster et al., 2022). Soil shear strength resulting from the sliding of soil particles during the penetration of is influenced by a combination of factors, including cohesion, friction, growing roots (Marshall et al., 1996). Soil penetration resistance, thus, and effective stress. These factors interact to determine how resistant the * Corresponding author. E-mail address: michael.adu@ucc.edu.gh (M.O. Adu). https://doi.org/10.1016/j.still.2023.105811 Received 21 January 2023; Received in revised form 17 June 2023; Accepted 24 June 2023 Available online 1 July 2023 0167-1987/© 2023 Elsevier B.V. All rights reserved. F. Kumi et al. S o i l & T i ll a g e R e s e a r c h 233 (2023) 105811 soil is to penetration. Soil strength is a complex property influenced by for feeding ruminants and also has the potential as a biofuel feedstock numerous factors, including soil composition, structure, moisture con- for power generation. Vetiver is an evergreen, gramineous, and peren- tent, density, stress history, and vegetation. The interplay of these fac- nial grass exploited in soil erosion prevention and rehabilitation of tors affects the soil’s penetration resistance. Plant roots potentially metal-polluted soils. Its roots are a valuable source of commercial and influence soil strength by increasing the soil’s shear strength either medical essential oils (Chou et al., 2016). directly through mechanical resistance and anchorage or indirectly through transpiration-induced soil water loss. 2.2. Growth conditions On the other hand, the development of plant roots is significantly affected by the biological and physicochemical properties of the soil The experiments were conducted in the greenhouse of the Teaching (Bengough et al., 2011). Increased mechanical impedance, induced by and Research Farm of the School of Agriculture, the University of Cape soil drying or compaction, affects soil porosity, all aspects of root Coast (UCC), Ghana. It was conducted under the natural day length of growth, rooting depth, and root system morphology. It also affects plant the area, ranging from approximately 11.30 to 12.40 h. During the nutrient and water availability and uptake functions and soil microbial experimental period, conditions in the greenhouse were: active photo- activities, effectively causing a decline in crop growth, productivity, and synthetic radiation of 486 ± 135 µmol m-2 s-1 over the plants for the yield (Mamo and Bubenzer, 2001; Raper and Mac Kirby, 2006; Whalley 11.30–12.40 h per day, air temperature within 33.5 ± 9.1 oC and rela- et al., 2008). There could, therefore, be a reciprocal relationship be- tive humidity of 62 ± 14% daily. Three soil types with varying textures tween soil strength and root system growth and functioning. were used for this study to allow for the assessment of the consistency of A typical soil mechanical property for assessing soil strength is soil plant species’ effect on soil penetration resistance or cone index (CI) penetration resistance or soil cone index (CI) measured by a cone across soil types. The soils were topsoil (0 − 15 cm depth) of two sandy penetrometer. The cone penetrometer is a handy and easy-to-use device clay loam soils of the Edna-Bronyibima/Benya-Udu series Acrisol for measuring soil strength in cohesive and non-cohesive soils (Amer- (Schad, 2016) (5.1294◦N, 1.2857◦W; called hereafter UCC soil), and atunga et al., 2016; Okello, 1991). A data logger records the force per Adawso-Bawjiase/Nta-Ofin series Forest Ochrosol (Schad, 2016) area unit required for the cone to press through the soil profile. The (5.1826◦N, 1.2485◦W; called hereafter as JUKWA soil), both typical recorded force per area unit is the CI (ASAE standard (S313.3), 2018; arable soils of Ghana’s coastal savannah agroecological zone. The last Hall and Raper, 2005). Senneset and Janbu (1985) suggested that cone growth media was beach sand, named hereafter SEA sand, which was penetration tests may obtain effective shear strength parameters, given also collected at 0 − 15 cm depth from the littoral zone of Cape Coast in that the CI estimates tip resistance related to the undrained shear the Central region of Ghana (5.0985◦N, 1.3109◦W). Before use, the strength of saturated, cohesive soil. The CI also estimates the sleeve beach sand was desalinated, and its electrical conductivity (EC) was friction associated with the friction of the soil horizon being penetrated determined to ensure it met the salinity requirements of the crop plants. by the cone penetrometer. Thus, the CI indicates soil strength and is a The EC, measured with Hanna pH/EC meter H15222 (Hanna Instrument compound parameter involving soil shear, compressive and tensile Inc., USA), was 0.075 dS/m. Since the beach sand had become more or strength, and soil-metal friction (Mulqueen et al., 1977). Factors less inert, sufficient nutrients were supplied to all its pots through fer- affecting CI include soil sampling depth, texture, moisture content, soil tigation with a modified Hoagland solution during plant growth. organic matter, bulk density, electrical conductivity, and cropping (Abbaspour-Gilandeh et al., 2012; ASAE Standard (EP542.1), 2019). 2.3. Determination of soil physicochemical properties Although many studies have been done to understand how soil strength affects the development of root systems (E.gs., Bengough and The soil’s field capacity (FC) was determined using the gravimetric Mullins, 1991; Ferreira et al., 2022; Willatt and Sulistyaningsih, 1990), method (Cassel and Nielsen, 1986). Briefly, plots measuring 1 m by 1 m more is needed to know about the feedback effects of root-induced from two locations, i.e., Aseibu-Nkrofro (Jukwa) and UCC, were flooded changes on soil strength. Earlier works by Mamo and Bubenzer (2001) and then covered with a thin plastic sheet to reduce evaporation. The found that soils with roots had twice the shear strength as those without plastic sheet covering was left for 2–3 days to allow for saturation and roots. Maffra et al. (2019) also confirmed that roots increase soil shear free drainage of gravitational water. On the third day, core samples were and compressive strengths. However, the authors noted that the kind of taken at 0–20 cm depth in three replicates from each plot’s centre, and contribution of roots to soil strength for clayey soil is different from the weight of the soil and core sampler were noted. The sample was then sandy soils. The question raised but not often considered is whether oven-dried at 105 ◦C until constant weight. Gravimetric water content roots could increase soil strength in different soils, whether plant species was determined by computing the difference between wet and dry could impact soil strength differently, and the interaction between masses divided by the mass of the dry sample. The bulk density of the various crop plant species and soil types. Therefore, the objective of the Aseibu-Nkroful soil, hereafter called JUKWA soil and UCC soil, was present study was to quantify the effects of roots from different crop estimated as the dry mass weight of the soil core divided by its volume. genetic materials on CI as a proxy for soil strength of varying soil types. The beach sand’s FC was determined using a modified hydrogel We hypothesized that plant roots impact CI, and root-induced soil method (Akhter et al., 2004). This method was adopted because the strength differs based on crop genotype and soil texture. beach sand was already taken from its source and had significantly been disturbed; its particle sizes were bigger, similar to the hydrogels used in 2. Materials and methods Akhter’s study. Briefly, cylindrical metallic containers about 10 cm high and 10 cm wide were perforated at the base and subsequently lined with 2.1. Genetic material filter paper and weighed. The metallic container with lined filter paper (hereafter referred to as the set-up) was placed on a weighing scale and In this study, four crop plants of the grass family, primarily cultivated tarred before it was filled with SEA sand by gently tapping at a pre- for food or fodder, namely maize (Zea mays L. cv. Obatanpa), sorghum defined frequency. A 1-cm headspace was left to the brim at the end of (Sorghum bicolor L. cv. Naga red), Napier grass [Cenchrus purpureus each filling to allow for adequate water saturation. The entire set-up was (Schumach)] and vetiver (Vetiveria zizanioides) were used. Maize is a then placed in a bowl with some volume of water, just enough to allow widely grown crop in 75 countries (Shiferaw et al., 2011), and in Ghana, soil saturation and avoid flooding. After saturation, the set-up was it accounts for 50% of the total cereal production. Sorghum is a staple removed from the bowl and gently placed on a cloth for free drainage. food for over half a million people, mainly in Africa, and is used in the The surface of the set-up during the free drainage period was covered bioethanol, fuel, brewing, sugar, or syrup industries (Kumar and Dwi- with a plastic film. Subsequently, the set-up was weighed and oven-dried vedi, 2020). Napier grass is a high-yielding and nutritious grass species at 105 ◦C until constant weight. Water content and soil bulk density 2 F. Kumi et al. S o i l & T i ll a g e R e s e a r c h 233 (2023) 105811 were then computed as was done for the JUKWA and UCC soils. The FC respectively. For each crop, these times marked the period of anthesis. was determined as the difference between the wet and the oven-dried Although the target was to excavate each crop at anthesis, this was soil divided by the wet soil. The bulk density of the SEA sand was impossible for Napier and vetiver grass. Napier grass is a perennial grass determined as the weight of the set-up and the sample volume. lasting 3–5 years. While it is ready for harvest 3–4 months after planting, The soil texture was determined using the pipette method following the mean days to flowering can be as high as 240 days after emergence the procedures of Anderson and Ingram (1994). The pH of the soils was (Sinche et al., 2021). Here, the roots of the Napier grass reached the end determined in water using soil-to-water ratio suspension of 1: 2.5 (w/w of our 70 cm columns by 70 days after planting. Similarly, vetiver is a basis) after agitation for 5 min and allowed to stand for additional 2 min. perennial grass whose flowering time can be extended. Some cultivars The measurement used a pH meter with a cross-bridge electrode (Hanna recorded flowering 240 days after planting (Suleiman et al., 2018). Instrument Inc., USA). The organic carbon was determined using the Accordingly, Napier and Vetiver grasses were excavated 70 days after dichromate method of Walkley and Black (Walkley and Black, 1934). emergence but had not yet reached anthesis. The experiment was con- Total nitrogen (TN) determination of soils was done using the micro ducted two consecutive times. The first experiment was planted on 7th Kjeldahl method after the digestion of soils in sulphuric acid with sele- March 2022, and the second was planted on 11th April 2022. nium powder as a catalyst (AOAC, 1990). The modified Molybdenum Blue method determined available phosphorus (AP) in soils (Murphy and Riley, 1962). The buffered ammonium acetate extractant was used 2.6. Measurements in extracting the exchangeable bases (Ca2+, Mg2+, and K +). Measure- ment of Potassium (K) was done by flame photometry (Jenway PFP 7 2.6.1. Cone penetration resistance model, Fischer Scientific, Goteborg, Sweden), and Ca and Mg were Soil moisture was measured with Acclima Digital True TDR-315 H measured using AAS (Buck Scientific model 210 VGP, Norwalk, USA). Sensor (Acclima, Inc. Idaho USA) during the experiment for irrigation The measured physicochemical properties of the three soils are shown in purposes and before the measurement of cone penetration resistance. Table S1. We employed a manually operated static penetrometer for penetrometer tests. The penetrometer was an Eijkelkamp penetrologger CBR (model 2.4. Column experiment set-up 0615SA) with a 60◦ circular steel cone and a base area of 100 mm2 and can measure up to a depth of 80 cm. The shoots of the plants growing in We used plant growth columns made of polyvinyl chloride (PVC) the columns were cut close to the soil surface to enable the depth pipes (75 cm high, 20 cm internal diameter) for the experiment. The reference plate to lay flat on the soil within the column. The pene- columns were lined with removable polypropylene woven sacks (pp trometer was then operated by placing the cone on the soil surface with woven sacks) sewed to fit to facilitate the removal of the soil cylinder the shaft oriented vertically and within the circular hole of the depth and excavation of roots. The bottom of the polypropylene woven bags reference plate. As much as possible, the depth reference plate was was sown, and the drawn-out tops were folded over the PVC columns’ placed so that its circular hole avoids the shoot stump left on the soil top end so they could be pulled out with the bulk soil at harvest. The surface. An operator exerted a force to push the rod into the soil. bags were permeable to water. The bags within columns were packed Penetration resistance of respective soil columns was measured at with each soil type up to the height of 70 cm to the bulk density that depths of 0–25, 25–40, 40–55, and 55–70 cm. The pressure applied to mimics the field dry bulk density of respective soils. Accordingly, pots of the cone or the force was normalised to the basal area of the cone to form the UCC, JUKWA, and SEA soils were filled to a dry bulk density of 1.3, a parameter called the CI. The difference between the CI of rooted and 1.3, and 1.7 g cm-3, respectively. All columns were filled with air-dried non-rooted pots was computed to estimate roots’ contribution to soil soil and sieved to remove aggregates and vegetative materials. The soil strength. packing into columns was performed in layers with loosening and roughing of the surface between layers to allow for a good connection 2.6.2. Plant growth and root system between packed layers and to avoid displacement of fine soil particles. At harvest, root-permeated and control soils were removed from the We covered the top of the columns with a layer of fine gravel to avoid columns by pulling the soil-containing polypropylene woven bags from surface effects from irrigation water and to reduce evaporation. the PVC columns. The soil cylinder was divided into four depths by slicing with a saw while still in the polypropylene woven bags. The 2.5. Experimental design and plant establishment upper part was 25 cm, and the following three parts were cut in 15 cm increments. We tagged the slices and moved them in a wheelbarrow to a The experiment was conducted using a factorial arrangement in a washing station. The polypropylene woven bags were gently severed completely randomised design. Four replicate pots per crop plant per longitudinally on both sides to remove them from the soil. The soil was soil type and four control pots unplanted for each soil were included. then submerged in water for approximately 1 h. The roots were gently Maize and sorghum were sown from two seeds per pot. After germina- removed and washed with tap water from a pressurized hose. The roots tion, the seedlings were thinned to one plant per pot. Napier grass was were gently shaken before being placed in a bowl of clean water, where planted with cuttings, and vetiver grass was planted from slips, each they were brushed and washed to remove any last bits of soil and other with one propagule per pot. For Napier, 3-node canes were cut and debris. buried in a slanting position in the columns, ensuring that the soil The cleaned roots and a scale object were placed on a black matte covered at least two nodes. For vetiver, plants were cut 25–30 cm above background to take images with a Canon EOS 70D DSLR camera, sus- the ground and dug out, and the culms were divided into slips with 2–3 pended with a tripod 0.6 m above the roots. Data were determined from tillers. We used a dibber to create a 3 cm deep hole in the soil. The slips the root images using image analysis software, Rhizo Vision Explorer, were pushed into the holes like seedlings, and the soils around the plants version 2.0.3, using the “Broken Root” mode (Seethepalli and York, were pressed. Except for SEA sand pots which received nutrient solu- 2020). The image thresholding level, edge smoothing, and root pruning tions as stated earlier, all other pots were watered every second day with threshold options were set at 230 pixels intensity, 2, and 1 pixel, tap water to 70% FC, initially determined gravimetrically. Subse- respectively. Pixels were converted to SI units (mm). Root features such quently, a soil moisture sensor (Acclima Digital True TDR-315 H Sensor; as total root length, branching frequency, average diameter, and volume Acclima, Inc. Idaho USA) was used to determine how much water to add were extracted. Finally, shoot and root dry matter was measured after at every irrigation event. oven-drying at 70 ◦C until constant weight. The root length density The duration of the plants’ growth was between 60 and 70 days. (total length of roots per unit of soil volume; RLD) was estimated for Maize and sorghum were excavated 60 and 70 days after emergence, each depth. 3 F. Kumi et al. S o i l & T i ll a g e R e s e a r c h 233 (2023) 105811 2.7. Data analysis unplanted control. The CI of soils grown to sorghum (1.5 MPa for both experiments) and vetiver grass (1.3 and 1.4 MPa) were similar, but both Statistical analyses were performed using the R software program were higher than the CI of the soil in the unplanted control columns (1.1 version 3.6.1 (R. Core Team, 2019). The Shapiro-Wilk test was con- and 1.0 MPa). The CI of the SEA sand (~2.0 MPa) was between 20% and ducted to verify the normality of data assumption. The CI data were 41%, significantly higher than that for the other two soil types (Fig. 1). logarithmically (ln) transformed as they had log-normal distribution. When the CI of the control pots is subtracted from that of root- Analysis of variances (ANOVA) was performed, first for the bulked data permeated pots, the difference could be ordered as follows: Napier and then for individual crop or soil type, to determine whether crop and grass>maize>sorghum>vetiver grass (Fig. 2a). The difference for soil type significantly affected the variables measured in experiments 1 Experiment 2 for each crop plant was higher (Fig. 2a) and varied among and 2. The factors for the ANOVA were plant species and soil type for the tested soils (Fig. 2b), higher for the SEA sand for the plant species. root system traits but also included soil depth in the analyses of CI and The difference decreased with increasing soil depth in the first experi- RLD. When ANOVA detected a significant difference among group ment but not in the second (Fig. 2c). Although CI appeared inversely means, multiple comparisons were performed using Tukey’s HSD to find related to soil water content, particularly for the second experiment, it the significantly different means. Pearson correlation analysis deter- was poorly associated with gravimetric soil water content measured at mined the relationship between the CI and measured crop parameters. 0–20 cm in both experiments (Fig. 2d). The significance criterion for all statistical analyses was p < 0.05. 3. Results 3.2. Root system architecture traits of the investigated plants 3.1. Soil penetration resistance Total root length showed significant differences (p < 0.001) among the crop plants and soil types (Fig. S1). Root length density (RLDs) In both experimental periods, crop plants, soil types, and soil depths decreased down the soil profile, generally peaking at 40 cm and significantly affected soil CI (Table S2). The CI generally increased with declining in the layers below 40 cm (Fig. 3a-f). The densities in the top increasing depth (Fig. 1). The overall CI of soils grown to Napier grass 25 cm and 40 cm ranged from 0.10 to 2.56 cm cm-3 and (2.0 and 2.3 MPa) and maize (1.7 and 2.2 MPa) were similar, but both 0.11–4.00 cm cm-3, respectively, and were variable between crops and were higher than the soils cultivated with the other crop plants and soils (Table S2). Vetiver and sorghum had lower overall RLDs than maize and Napier across the profile. In both experiments, RLDs in the SEA sand Fig. 1. Cone index measured at 0–70 cm depth for the control, maize, Napier, sorghum, and vetiver cultivated on JUKWA, SEA, and UCC soils in Experiment 1 (a–c) and Experiment 2 (d–e). The dotted line indicates the frequently-stated upper threshold value of soil strength where plant root growth can be impeded. Treatments without letters or the same letters at a given depth are not significant at p < 0.05. 4 F. Kumi et al. S o i l & T i ll a g e R e s e a r c h 233 (2023) 105811 Fig. 2. a-c: Difference in cone index between unplanted control and root-permeated pots a) Difference for the four plant spps.; b) the difference between the three soil types; c) the Difference at various depths; and d) linear regression between cone index and soil moisture content at 0–20 cm depth. were the lowest (Fig. 3a-f). The distribution of branching frequency per 3.4. Plant parameters and cone index relationships mm, average diameter, and volume of roots for the plant species was highly variable, but some general trends were observed. The branching The correlation between mean CI measured at 0–70 and root pa- frequency of roots for the plants generally increased with depth rameters were highly variable for the different plants and soils investi- (Table 1). In contrast, the average diameter of the root generally gated. All but one of the significant correlations between RLD and CI decreased with depth (Table 1). Overall, maize had the largest average occurred in vetiver grown in the SEA sand (Table 2). In the first exper- root diameter, particularly at 0–25 and 25–40 cm depth. Also, the maize iment, the RLD of Napier grass was strongly correlated with CI in the plant tended to have the highest root volume compared to the vetiver, JUKWA (Table 2). In the second experiment, the RLD of vetiver grown in which had the lowest root volume, particularly for the JUKWA (Table 1). the SEA sand was negatively correlated with CI for maize and Napier. SEA. Still, the RLD of sorghum and vetiver was positively correlated with CI (Table 2). The CI generally negatively correlated with average plant root 3.3. Shoot and root biomass of crops diameter, except for vetiver grass for UCC in Experiment 1 and vetiver for JUKWA in Experiment 2 (Table 2). For example, for both experi- There were significant differences (p < 0.001) in shoot dry weight ments using the SEA soil, it was observed that the CI tends to be nega- (SDW) between the crop plants and soil types in both experiments tively correlated with the root diameter of maize and Napier. In the case (Table S2). The SDW among the crop plants could be ordered as follows: of the JUKWA soil, the CI correlated negatively with branching fre- Napier > maize = sorghum > vetiver, and among the soil types was quency for maize in the first experiment but showed a reverse trend for JUKWA soil = UCC soil > SEA sand (Fig. 4a-c). Both experiments had the second experiment. The UCC soil correlated positively with significant (p = 0.006 and p = 0.016) crop x soil interaction (Table S2). branching frequency for Napier, while a negative correlation was Overall, root biomass significantly (p < 0.001) differed between crop observed for root diameter in both experiments. plants, soil types, and soil depth in the first experiment (Table S2). The root biomass of Napier grass (4.7 g) was about 2-, 3- and 5-fold higher 3.5. Soil moisture content than that of maize (3.0 g), sorghum (2.0 g), and vetiver (1.0 g), respectively. The root biomass of plants grown in the JUKWA (3.4 g) and Overall, the moisture content at 0–20 cm depth of the three soil types UCC (3.3 g) soils were comparable, and these were about 3-fold higher was significantly different (p < 0.01) at the time of excavation. There than that of the plants grown in the SEA sand (1.1 g). Root biomass was was about a 31% difference in the moisture content of the soils in the concentrated in the surface soil (0–25 cm depth) for all crop plants and first (0.12 kg kg-1) and second (0.17 kg kg-1) experiments (Table S2; decreased with increasing soil depth (Fig. 4d-i). Fig. 5). 5 F. Kumi et al. S o i l & T i ll a g e R e s e a r c h 233 (2023) 105811 Fig. 3. Root length densities at different depths of maize, Napier, sorghum, and vetiver cultivated in JUKWA, SEA, and UCC soils during Experiment 1 (a-c) and Experiment 2 (d-f). Different letters indicate a significant difference between crops for a given soil depth at p < 0.05. 4. Discussion determine the roots’ contribution to CI and soil strength (Fig. 2), but this approach may be overly crude. It fails to determine precisely how much 4.1. Contribution of RSA to cone index of the difference between pots with roots versus pots without roots is due to the roots’ reinforcement and how much is due to differences in water The central hypothesis of the present study was that plant roots content. Even so, the approach offers a starting point to disaggregate the provide additional confining pressure to the soil and contribute to the increased rooted soil strength factors. soil’s penetration resistance, which is an indicator of soil strength. Still, Contrary to that observed by Lardy et al. (2022) and Zhang et al. the foremost question needing answering is whether soil penetration (2006), CI did not vary linearly with gravimetric water content, sug- resistance indicates soil strength. Soil penetration resistance is an gesting that the contribution of soil moisture content to CI in the present essential indicator of soil strength (Jiang et al., 2020; Weaich et al., study was minimal. Instead, soil bulk density and roots’ influence on 1992). Soil penetration resistance estimates the cohesive forces between cohesion and internal friction might govern the increased CIs and soil soil particles and frictional resistance met by an inserting object (Cairns strength of root-permeated soils in the present study. Assessing the root et al., 2011). It, therefore, feeds into the significant computational pa- tensile strength and soil shear strength of root-permeated soils will be rameters of soil strength. Thus, root penetration resistance, measured as critical in quantifying the direct effects of roots on soil strength. Liang CI in the present study, provides a suitable proxy for soil strength et al. (2020) pointed out that the strength provided by roots is a function because CI estimates tip resistance which is theoretically related to the of their properties, such as tensile strength and root density. The lack of undrained shear strength of saturated and cohesive soils. The CI also a significant inverse relationship between CI and soil moisture was un- estimates the sleeve friction, which is theoretically associated with the expected and might be because all pots were irrigated similarly in the friction of the soil horizon being penetrated. present study. Although slight differences in soil moisture content might In this study, the root-embedded soils presented higher CIs. Internal have emerged for the different crops due to their differences in water friction and cohesion are the two main important factors that determine uptake by the plants, it was insufficient to cause a significant relation- soil shear strength; in sandy soil, roots can increase by 234% (Maffra ship between soil moisture and CI. et al., 2019). Thus, increasing cohesion may be implicated in the high CI It is also important to note that although sandy soils typically have of the root-permeated soils. Increased CI of the root-permeated soils weak cohesion forces among particles, they also tend to have very high could be attributed to reduced soil moisture occasioned by water uptake values of internal angle friction due to their large particle size. This by roots and reconsolidation due to increased soil bulk density. We increases their CIs and explains the high values recorded for the SEA computed the difference in CI between control and rooted pots to sand in this study. Previous studies examining the relationships between 6 F. Kumi et al. S o i l & T i ll a g e R e s e a r c h 233 (2023) 105811 Table 1 Branching frequency, average diameter, and roots volume at 0–25, 25–40, 40–55, and 55–70 cm depth for maize, Napier, sorghum, and vetiver cultivated on JUKWA, SEA, and UCC soil during Experiments 1 and 2. Different letters indicate a significant difference between crops for a given soil depth at p < 0.05. Soil type Depth (cm) Branching frequency per mm Average diameter (mm) Volume (mm-3) Maize Napier Sorghum Vetiver Maize Napier Sorghum Vetiver Maize Napier Sorghum Vetiver Experiment 1 JUKWA Soil 0–25 0.593a 0.632a 0.490b 0.454b 1.914a 1.599a 2.049b 1.616b 295951a 134767bc 176422ab 32558c 25–40 0.636a 0.665a 0.642a 0.473b 1.682a 1.642ab 1.212b 1.406ab 202221 180111 39303 32176 40–55 0.643a 0.642a 0.666a 0.489b 1.481a 1.524a 1.146b 1.388ab 141960ab 214794a 25633b 36984b 55–70 0.674a 0.666a 0.685a 0.469b 1.452 1.465 1.301 1.315 166255 191473 73208 30466 SEA Soil 0–25 0.59 0.81 0.56 0.52 1.91 1.84 1.89 1.65 206248 132880 83691 44225 25–40 0.66a 0.67a 0.70a 0.48b 1.59 1.38 1.09 1.38 103515 94350 6836 18549 40–55 0.63ab 0.71a 0.74a 0.52b 1.60a 1.04b 1.07b 1.65a 96461a 8382b 8730b 32019ab 55–70 0.69ab 0.77a 0.82a 0.55b 1.22ab 0.98b 0.97b 1.45a 23814 12673 7065 27056 UCC Soil 0–25 0.77 0.95 0.49 0.72 2.11 2.51 1.92 2.47 176481 302254 105934 104709 25–40 0.87 0.82 0.60 0.71 1.59 1.74 1.37 2.50 127249 153982 51203 221351 40–55 0.91 0.69 0.63 0.77 1.55 1.34 1.09 1.99 90413 97553 25301 38189 55–70 0.87 0.72 0.72 0.66 1.76 1.20 1.35 2.15 125622 56488 32439 14767 Experiment 2 JUKWA Soil 0–25 0.380b 0.437b 0.376b 0.495a 2.31a 2.66a 2.28a 1.50b 512502b 1396277a 262256b 118360b 25–40 0.443 0.460 0.512 0.491 1.67b 2.18a 1.42c 1.49bc 252586b 869805a 115691b 108137b 40–55 0.444 0.459 0.551 0.547 1.92a 2.02a 1.31b 1.31b 366337ab 647996a 103741b 130597b 55–70 0.479ab 0.471b 0.549a 0.493ab 1.84ab 2.05a 1.41b 1.45b 374523ab 666496a 154168b 131797b SEA Soil 0–25 0.406 0.574 0.467 0.409 2.20 2.68 2.11 1.91 422484 813224 305547 121538 25–40 0.470b 0.473b 0.565a 0.411b 2.06a 1.50b 1.56b 1.53b 268440 150446 149943 54573 40–55 0.485 0.391 0.610 0.423 1.81 1.05 1.40 1.84 72770 81468 190246 95962 55–70 0.534 0.392 0.553 0.444 1.47 1.11 1.68 1.64 79099a 115569ab 224267ab 69324b UCC Soil 0–25 0.456 0.409 0.429 0.430 2.24 2.53 2.27 1.55 1029059 1243016 423952 91378 25–40 0.531 0.459 0.564 0.425 1.96 2.14 1.66 1.51 912976 925814 433841 124712 40–55 0.531ab 0.476ab 0.594a 0.437b 1.93ab 2.28a 1.35b 1.43ab 803684 1510859 267607 67400 55–70 0.531 0.479 0.568 0.476 1.76 2.12 1.63 1.54 510021 1028371 444812 67053 root traits and mechanical soil properties have shown that the RLD the root system’s contribution to soil strength and stabilisation. Signif- influenced soil shear strength. In that context, the plants with high RLD icant genetic variations were observed for the RSA traits evaluated, present increased soil strength. In the present study, the RLD of Napier including biomass, length and RLD, diameter, and the number of roots grass was strongly correlated with CI for the JUKWA Soil, and the RLD of (Figs. 1 and 2 and Tables S2 and 1). Napier and maize generally had sorghum and vetiver was positively correlated with CI (Table 2). Even higher root biomass and sizes than sorghum and vetiver for all investi- so, the RLD of the four plant species did not establish a consistent gated soils. However, Napier was superior up to 40 cm depth, agreeing relationship with CI. The lack of consistency could be explained by the with Sekiya et al. (2013) that Napier grass possesses more significant fact the CI measures penetration resistance not only due to the presence root systems than other grasses. The extensive root systems of the Napier of roots but also due to other factors. The low RLD of the four investi- were also reflected in its above-ground biomass, recording the highest gated plant species (0.003–4.00 cm cm-3) compared to the RLD shoot biomass across both experiments (Fig. 1). As reported by several measured for perennial herbaceous, shrub, and tree species (more than others (e.gs., Angima et al., 2002; Dos-Santos et al., 2021), the Napier 300 cm cm-3) might explain the absence of significant relationships grew very fast and vigorously, its roots reaching the bottom of the 70 cm between RLD and the CI for some of the plant species in the present column in just about a month. Thus, the present study suggests Napier study. grass develops fast-growing and extensive root systems with substan- Several processes through which root systems might influence soil tially large biomass, lengths, RLD, and numbers. Napier, therefore, strength have been highlighted in previous reviews (Angers and Caron, represents a possible plant to explore aspects of nature-based soil 1998; Loades et al., 2013). For example, roots growing in existing pores management, including enhancing soil strength and stabilisation. can create compressive shear stresses reaching 2 MPa (Goss, 1991), akin In all cases, the vetiver recorded the least values, an occurrence that to the CI measured in this study. Where the size of the root system is its slow establishment in the present study may explain. The vetiver’s small in the ground, the amount of water loss through transpiration poor performance here is somewhat perplexing. Typically, vetiver is a might be equally small, probably explaining why the soil moisture fast-growing, stress-tolerant plant whose extensive root system can content of columns sown to vetiver was similar to the control columns reach a rooting depth of 60 cm in 3 weeks (Yoon, 1991) and 3–4 m in the (Fig. 5). The aggregate binding of the soil by root can be through first year (National Research Council, 1993). Perhaps it is worth enmeshing with a network of roots or root exudation and mucilage. repeating that this work was performed in unnatural environments Plant root hairs of many crop plants form rhizosheaths (Adu et al., 2017; involving soil-filled columns and a greenhouse. Despite its tolerance to Opoku et al., 2022) and are critical in enmeshing and stabilising soil stressful conditions, vetiver grows best in the open and has been noted to aggregates. It is instructive to determine how root hairs of the tested be highly intolerant to shading, which is said to reduce its growth crop plants contribute to soil strength and stabilisation. markedly (Truong and Baker, 1998). Suboptimal photosynthetically active radiation for vetiver and column confinement might have thus 4.2. Implication for genotypic variation in root system traits to cone index adversely impacted the vetiver’s establishment in the present study. In the long term, vetiver is a deeply rooted, persistent grass that offers a According to Abdullah et al. (2011), root systems of different crop practical and inexpensive solution for controlling erosion when planted plants influence soil strength differently, corroborating the observations along the contours of sloping lands (National Research Council, 1993). in this work. Root parameters such as diameter, volume, biomass, In bioengineering, an integrated system of Napier and vetiver will take numbers, and length have been noted to play essential roles in modi- advantage of the rapid establishment of Napier until the vetiver is fully fying soil structure and strength (Foresta et al., 2020; Kumi et al., 2016), established. However, Napier roots’ tensile strength, root decay and and therefore, genotypic variation in these traits would be reflected in resultant soil shear strength must be validated. 7 F. Kumi et al. S o i l & T i ll a g e R e s e a r c h 233 (2023) 105811 Fig. 4. a-c: Shoot dry weight of maize, Napier, sorghum, and vetiver grown on the (a) JUKWA, (b) SEA, and (c) UCC soils investigated during Experiments 1 and 2. Error bars indicate the standard error of the mean. Treatments with different letters are significantly different for a given soil at p < 0.05. Upper and lower case letters denote Experiments 1 and 2, respectively; d-i: Root dry weight measured at 0–70 cm depth for the control, maize, Napier, sorghum, and vetiver cultivated on JUKWA, SEA, and UCC soils in Experiment 1 (d-f) and Experiment 2 (g–i). Different letters indicate a significant difference between crops for a given soil depth at p < 0.05. 4.3. Effect of type of soil on RSA and cone index with plant species (Loades et al., 2013). The average soil strengths across the two experiments for the unplanted SEA sand, JUKWA, and UCC soils Plant roots grow unimpeded when soil strength is below 0.7 MPa, were 1.3, 0.8, and 0.6 MPa, respectively, indicating that root growth but root growth is directly affected between 0.7 MPa and 2.0 MPa. At may have been impeded in the SEA sand. The high soil strength of the soil strength of 2 MPa or more, mechanical penetration of roots is seri- SEA sand might be due mainly to the higher sand content, its attendant ously hindered (Barnhisel and Hower, 1997), but the threshold varies high bulk density (1.80 g cm-3: Table S1), and its limited capacity to 8 F. Kumi et al. S o i l & T i ll a g e R e s e a r c h 233 (2023) 105811 retain soil moisture for plant uptake (Fig. 5). Again, the high penetration resistance recorded in the SEA sand could be due to an increased angle of internal friction due to the high particle or packing density of this type of soil. Typically, drier soils tend to be more compact, which tends to resist the growth and development of roots (Fan and Su, 2008; Kumi et al., 2016; Raper and Mac Kirby, 2006). Typical sandy soil has very low cohesion but high internal shearing resistance, leading to reduced root growth (biomass, root length, and numbers) and spatial distribution (Figs. 4 and S1). Cell division and size in mechanically impeded roots can decline by 40% (Bengough and Mullins, 1990). Compared to the arable soils (i.e., the JUKWA and UCC soils), the total root length in the 70 cm soil col- umns reduced, between 48.3% and 57.3% when plants were grown in the SEA sand. Indeed, the reduced root growth cannot be attributed only to penetration resistance in the SEA sand but also to other factors, such as possible declined soil fertility and moisture in the SEA sand, as stated previously. Even so, the present results suggest that the effect of soil strength on cell division and growth is also evident in other root systems traits, such as the total root length of matured plants, and the effect may vary with plant species. When maize was planted in nine different soils, differences were observed in the root biomass and other RSA traits (Böhm, 1979). Loades et al. (2013) showed that under increased soil penetration resistance, root growth rate might be impeded by the structural degradation of soil pore spaces, with attendant decreases in porosity, hydraulic conductivity, and air permeability. 4.4. Effect of depth on RSA and cone index The results in the present study corroborate the reports that CI usually varies with penetration depth (Wismer and Luth, 1974). The CI generally increased with increasing depth (Fig. 1), with a 57–99% in- crease in CI of the bottom layer compared to the top layer of the soil column. The increased CI with depth was, possibly, in response to increasing effective confining stress and increased cohesion of soil par- ticles with depth. Therefore, as Zhao et al. (2018) pointed out, roots’ contribution to the overall shear strength decreases with increasing depth. In the present study, root branching frequency generally increased with depth, but the opposite was true for average root diam- eter (Table 1). Thinner roots could grow through soil pores to deeper layers and produce laterals (Hendry et al., 2014; Putri et al., 2010; Zhao et al., 2018). On the other hand, the growth of larger root axes may be impeded, especially in compacted and drier soils. Soil strength on root distribution and natural variability in root depth between species appear to influence the variability in root system distribution with depth. For example, Napier, maize, and sorghum partitioned a considerable amount of root biomass in the upper layer (Fig. 3). Still, in both experiments, root biomass and length changed only slightly across layers for vetiver regardless of marked increases in CI with depth (Fig. 3 and 4). Napier, maize, and sorghum performed as typical grasses with fibrous root systems, which proliferate and partition in shallow soil layers, potentially increasing surface soil reinforcement and reducing surface soil erosion (Baets et al., 2007; Gyssels et al., 2005). Indeed, Angima et al. (2002) noted that Napier grass’ effective- ness for erosion control could be due to its massive near-surface lateral root system. According to Wang et al. (2020), vetiver root system size decreases with increasing depth. The results demonstrated that if vetiver root systems are characterised by a reasonably constant proliferation and partitioning along the soil profile, it might offer a different or additional role in soil reinforcement compared to the other grasses. 5. Conclusions The study showed that roots of maize, Napier grass, sorghum, and vetiver contributed to an increase in soil strength quantified as cone index (CI), representing penetration resistance, but in varied ways. Although the CI provide a valuable and quick metric of soil strength, the 9 Table 2 Correlation Between Cone Index and Selected Root Parameters for Experiments 1 and 2. CI Experiment 1 0–70 cm depth Root length density (cm cm-3) Branching frequency per mm Average diameter (mm) Volume (mm^3) Soil type Maize Napier Sorghum Vetiver Maize Napier Sorghum Vetiver Maize Napier Sorghum Vetiver Maize Napier Sorghum Vetiver Pearson Correlation JUKWA 0.05 ns 0.99 * ** 0.32 ns 0.23 ns -0.95 * * 0.16 ns -0.74 ns 0.91 * -0.90 * -0.60 ns -0.03 ns -0.54 ns -0.74 ns 0.99 * ** * 0.05 ns -0.78 ns coefficient (r) Soil SEA Soil -0.43 ns -0.62 ns -0.83 ns 0.97 * * -0.92 * 0.12 ns -0.93 * 0.40 ns -0.90 * -0.93 * -0.95 * * -0.31 ns -0.86 * -0.78 ns -0.89 * -0.36 ns UCC Soil 0.70 ns -0.33 ns -0.39 ns -0.72 ns 0.68 ns 0.88 * -0.82ns 0.57 ns -0.93 * -0.98 * -0.32 ns 0.11 ns -0.95 * -0.93 * -0.55 ns -0.48 ns Experiment 2 0–70 cm depth Root length density (cm cm-3) Branching frequency per mm Average diameter (mm) Volume (mm^3) Soil type Maize Napier Sorghum Vetiver Maize Napier Sorghum Vetiver Maize Napier Sorghum Vetiver Maize Napier Sorghum Vetiver Pearson Correlation JUKWA 0.61 ns 0.65 ns 0.76 ns 0.41 ns 0.85 * -0.30 ns 0.82 ns 0.84 ns -0.32 ns -0.39 ns -0.55 ns 0.04 ns -0.05 -0.28 ns -0.23 ns 0.57 ns coefficient (r) Soil SEA Soil -0.88 * -0.97 * * 0.89 * 1.00 * ** -0.84 ns 0.94 * * 0.52 ns 0.06 ns -0.87 * -0.97 * * -0.73 ns -0.54 ns -0.99 * * -1.00 * * -0.43 ns -0.63 ns UCC Soil 0.66 ns 0.78 ns 0.42 ns 0.52 ns 0.62 ns 0.85 * 0.42 ns -0.91 * -0.94 * * -0.68 ns -0.34 ns -0.19 ns -0.80 ns -0.42 ns 0.24ns -0.79 ns ns: p > 0.05, * p ≤ 0.05, * * p ≤ 0.01, * ** p ≤ 0.001 and * ** * p ≤ 0.0001 F. Kumi et al. S o i l & T i ll a g e R e s e a r c h 233 (2023) 105811 Fig. 5. Soil moisture content at 0–20 cm depth during sampling for the (a) Jukwa, (b) SEA, and (c) UCC soils investigated during Experiments 1 and 2. Error bars indicate the standard error of the mean. Treatments with different letters are significantly different for a given soil at p < 0.05. Upper and lower case letters denote Experiments 1 and 2, respectively. estimates must be indexed with other factors since CI is strongly influ- K. Amoah, Kofi Atiah, Paul A. Asare: Investigation. Samuel Banafo, enced by soil texture, moisture content, bulk density and soil organic Michael O. Adu: Data curation. Michael O. Adu, Peter B. Obour, matter content. Here, simple linear regression suggested that soil Emmanuel Arthur, Francis Kumi, Kofi Atiah: Writing − original draft moisture content did not significantly increase the CI of root-permeated preparation. Michael O. Adu, Francis Kumi, Paul A. Asare, Kofi soils. The interaction of soil bulk density, root biomass, and morpho- Atiah, Kwadwo K. Amoah: Supervision. Michael O. Adu, Francis logical traits contributed substantially to soil strength. Plant roots may Kumi, Peter B. Obour, Emmanuel Arthur, Stephen E. Moore, Paul A. have impacted CI or soil strength by transferring soil shear stress into Asare, Joel Asiedu, Donatus B. Angnuureng, Kofi Atiah, Kwadwo K. root tensile resistance via the roots’ action on friction and interlocking Amoah, Shadrack K. Amponsah, Selorm Y. Dorvlo, Samuel Banafo: force between soil particles and the root system. As seen in the present Writing − review & editing. Shadrack K. Amponsah, Peter B. Obour: study, the transfer of soil shear stress into root tensile resistance might Provision of tools. Michael O. Adu, Francis Kumi, Peter B. Obour, be moderated by natural variability in root system traits between plant Emmanuel Arthur, Stephen E. Moore, Paul A. Asare, Joel Asiedu, species, the interaction between soil texture and root biomass, and root Donatus B. Angnuureng, Kofi Atiah, Kwadwo K. Amoah, Shadrack morphological characteristics. K. Amponsah, Selorm Y. Dorvlo, Samuel Banafo: Validation. Moreover, root traits and CI varied with increasing depth and for soil Michael O. Adu, Francis Kumi, Emmanuel Arthur, Paul A. Asare, types, indicating that the interaction of the physiological parameters of Stephen E. Moore, Joel Asiedu, Donatus B. Angnuureng: Funding plant roots, soil type, and depth of root profiles influences soil strength. acquisition. Overall, this short-term study showed that Napier grass, followed by maize, sorghum, and vetiver, presented superior soil penetration re- Declaration of Competing Interest sistances. A long-term study would be needed to ascertain how much Napier consistently can contribute to soil stability and strength. Such a The authors declare no conflicts of interest. study could include modelling the relationship between shear strength and angle of internal friction alongside root tensile strength and decay Data availability rate to provide an in-depth understanding of the underlying mechanisms that explain the contribution of the plant root system to soil-root Data will be made available on request. reinforcement. Acknowledgements Declaration of Funding We thank Dr David Oscar Yawson for his internal review and advice This research was funded by the Directorate of Research, Innovation, on the research and the manuscript. We would also like to thank Vincent and Consultancy (DRIC) of the University of Cape Coast (UCC), under Opoku Agyemang, Justice Asante, Godswill Hygienus, Azure Sanleri, the 6th Call for Research Support Grants. The study is part of the research and Solomon Amamu for their assistance in the lab work. We would also project with reference number: RSG/PAP/CANS/2021/101 and enti- like to thank the anonymous reviewers for their careful reading of our tled: “Exploiting the Root System Architecture of Indigenous Multi- manuscript and their many valuable comments and suggestions. functional Crop Plants for Nature-Based Erosion Management and Soil Reinforcement.” Appendix A. Supporting information CRediT authorship contribution statement Supplementary data associated with this article can be found in the online version at doi:10.1016/j.still.2023.105811. Michael O. Adu: Conceptualization. Michael O. Adu, Francis Kumi, Emmanuel Arthur, Peter B. Obour, Paul A. Asare, Stephen E. Moore, Joel Asiedu, Donatus B. Angnuureng, Kofi Atiah, Kwadwo K. Amoah: Methodology. Michael O. Adu, Samuel Banafo, Kwadwo 10 F. Kumi et al. S o i l & T i ll a g e R e s e a r c h 233 (2023) 105811 References Jiang, Q., Cao, M., Wang, Y., Wang, J., 2020. Estimating soil penetration resistance of paddy soils in the plastic state using physical properties. Agronomy 10, 1914. Abbaspour-Gilandeh, Y., Shaygani-Soltanpour, A., Golmohammadi, A., Rahimi- https://doi.org/10.3390/agronomy10121914. Ajdadi, F., Qahari-Kermani, F., 2012. Evaluation of factors affecting soil cone index Kumar, P., Dwivedi, P., 2020. Lignin estimation in sorghum leaves grown under and determination of mathematical model for prediction of soil cone index using hazardous waste site. Plant Arch. 20, 2558–2561. multiple-regression model. 8th Int. Soil Sci. Congr. “Land Degrad. Chall. Sustain. Soil Kumi, F., Mao, H., Li, Q., Luhua, H., 2016. Assessment of tomato seedling substrate-root Manag. ” Çemezmir, Turk. 325–329. quality using X-ray computed tomography and scanning electron microscopy. Appl. Abdullah, M.N., Osman, N., Ali, F.H., 2011. Soil-root shear strength properties of some Eng. Agric. 32, 417–427. slope plants. Sains Malays. 40, 1065–1073. Lardy, J.M., DeSutter, T.M., Daigh, A.L.M., Meehan, M.A., Staricka, J.A., 2022. Effects of Adu, M.O., Asare, P.A., Yawson, D.O., Ackah, F.K., Amoah, K.K., Nyarko, M.A., Andoh, D. soil bulk density and water content on penetration resistance. Agric. Environ. Lett. 7, A., 2017. Quantifying variations in rhizosheath and root system phenotypes of e20096 https://doi.org/10.1002/ael2.20096. landraces and improved varieties of juvenile maize. Rhizosphere 3, 29–39. Liang, T., Knappett, J.A., Leung, A., Carnaghan, A., Bengough, A.G., Zhao, R., 2020. Akhter, J., Mahmood, K., Malik, K.A., Mardan, A., Ahmad, M., Iqbal, M.M., 2004. Effects A critical evaluation of predictive models for rooted soil strength with application to of hydrogel amendment on water storage of sandy loam and loam soils and seedling predicting the seismic deformation of rooted slopes. Landslides 17, 93–109. growth of barley, wheat and chickpea. Plant, Soil Environ. 50, 463–469. https://doi. Loades, K.W., Bengough, A.G., Bransby, M.F., Hallett, P.D., 2013. Reinforcement of Soil org/10.17221/4059-PSE. by Fibrous Roots. In: Enhancing Understanding and Quantification of Soil–Root Ameratunga, J., Sivakugan, N., Das, B.M., 2016. Cone penetrometer test. In: Growth Interactions. John Wiley & Sons, Ltd, pp. 197–228. https://doi.org/ Ameratunga, J., Sivakugan, N., Das, B.M. (Eds.), Correlations of Soil and Rock 10.2134/advagricsystmodel4.c9. Properties in Geotechnical Engineering, Developments in Geotechnical Engineering. Maffra, C., Sousa, R., Sutili, F., Pinheiro, R., 2019. The Effect of Roots on the Shear Springer, India, New Delhi, pp. 115–157. https://doi.org/10.1007/978-81-322- Strength of Texturally Distinct Soils. Floresta Ambient. 26. https://doi.org/10.1590/ 2629-1_5. 2179-8087.101817. Anderson, J., Ingram, J., 1994. Tropical soil biology and fertility: a handbook of Mamo, M., Bubenzer, G.D., 2001. Detachment rate, soil erodibility, and soil strength as methods. Soil Sci. 157, 265. https://doi.org/10.2307/2261129. influenced by living plant roots. Part I: Laboratory study. Trans. ASAE 44, Angers, D.A., Caron, J., 1998. Plant-induced changes in soil structure: processes and 1167–1174. feedbacks. Biogeochemistry 42, 55–72. https://doi.org/10.1023/A:1005944025343. Marshall, T.J., Holmes, J.W., Rose, C.W., 1996. Soil Physics, 3rd ed..,. Cambridge Angima, S.D., Stott, D.E., O’neill, M.K., Ong, C.K., Weesies, G.A., 2002. Use of University Press,, Cambridge (UK). calliandra–Napier grass contour hedges to control erosion in central Kenya. Agric., Mulqueen, J., Stafford, J.V., Tanner, D.W., 1977. Evaluation of penetrometers for Ecosyst. Environ. 91, 15–23. measuring soil strength. J. Terra 14, 137–151. https://doi.org/10.1016/0022-4898 AOAC, 1990. Official methods of analysis. Association of Official Analytical Chemists. (77)90012-X. ASAE Standard (EP542.1), 2019. Procedures for Using and Reporting Data Obtained with Murphy, J., Riley, J.P., 1962. A modified single solution method for the determination of the Soil Cone Penetrometer. American Society of Agricultural and Biological phosphate in natural waters. Anal. Chim. Acta 27, 31–36. https://doi.org/10.1016/ Engineers. St. Joseph., Mich., USA 1. S0003-2670(00)88444-5. ASAE standard(S313.3), 2018, ASAE S313.3 Soil Cone Penetrometer -. National Research Council, 1993. Read “Soil and Water Quality: An Agenda for Baets, S.D., Poesen, J., Knapen, A., Galindo, P., 2007. Impact of root architecture on the Agriculture” at NAP.edu. https://doi.org/10.17226/2132. erosion-reducing potential of roots during concentrated flow. Earth Surf. Process. Okello, A., 1991. A review of soil strength measurement techniques for prediction of Landf.: J. Br. Geomorphol. Res. Group 32, 1323–1345. terrain vehicle performance. J. Agric. Eng. Res. 50, 129–155. https://doi.org/ Barnhisel, R.I., Hower, J.M., 1997, Coal surface mine reclamation in the eastern United 10.1016/S0021-8634(05)80010-1. States: The revegetation of disturbed lands to hayland/pasture or cropland. Opoku, V.A., Yawson, D.O., Asare, P.A., Afutu, E., Kotochi, M.C., Amoah, K.K., Adu, M. Bengough, A.G., Mullins, C.E., 1990. Mechanical impedance to root growth: a review of O., 2022. Root hair and rhizosheath traits contribute to genetic variation and experimental techniques and root growth responses. J. Soil Sci. 41, 341–358. phosphorus use efficiency in cowpea (Vigna unguiculata (L.) Walp). Rhizosphere 21, Bengough, A.G., Mullins, C.E., 1991. Penetrometer resistance, root penetration resistance 100463. and root elongation rate in two sandy loam soils. Plant Soil 131, 59–66. https://doi. Putri, E.E., Rao, N., Mannan, M.A., 2010. Evaluation of the modulus of elasticity and org/10.1007/BF00010420. resilient modulus for highway subgrades. Electron J. Geotech. Eng. 15, 1285–1293. Bengough, A.G., McKenzie, B.M., Hallett, P.D., Valentine, T.A., 2011. Root elongation, Raper, R.L., Mac Kirby, J., 2006. Soil compaction: how to do it, undo it, or avoid doing it. water stress, and mechanical impedance: a review of limiting stresses and beneficial Am. Soc. Agric. Biol. Eng. root tip traits. J. Exp. Bot. 62, 59–68. https://doi.org/10.1093/jxb/erq350. Schad, P., 2016. The international soil classification system WRB. 2014. Nov. Methods Böhm, W., 1979. Root parameters and their measurement. In: Methods of Studying Root Monit. Manag. Land Water Resour. Sib. 563–571. Systems. Springer, pp. 125–138. Seethepalli, A., York, L.M., 2020. RhizoVision. Explor. - Interact. Softw. Gen. root Image Cairns, J.E., Impa, S.M., O’Toole, J.C., Jagadish, S.V.K., Price, A.H., 2011. Influence of Anal. Des. Everyone. https://doi.org/10.5281/zenodo.4095629. the soil physical environment on rice (Oryza sativa L.) response to drought stress and Sekiya, N., Shiotsu, F., Abe, J., Morita, S., 2013. Distribution and Quantity of Root its implications for drought research. Field Crops Res. 121, 303–310. https://doi. Systems of Field-Grown Erianthus and Napier Grass. Am. J. Plant Sci. 2013. https:// org/10.1016/j.fcr.2011.01.012. doi.org/10.4236/ajps.2013.412A1003. Cassel, D.K., Nielsen, D.R., 1986. Field capacity and available water capacity. Methods Senneset, K., Janbu, N., 1985. Shear strength parameters obtained from static cone Soil Anal.: Part 1 Phys. Mineral. Methods 5, 901–926. penetration tests. Strength testing of marine sediments: laboratory and in-situ Chou, S.-T., Shih, Y., Lin, C.-C., 2016. Chapter 96 - Vetiver Grass (Vetiveria zizanioides) measurements. American Society for Testing and Materials,, Philadelphia, PA, Oils. In: Preedy, V.R. (Ed.), Essential Oils in Food Preservation, Flavor and Safety. pp. 41–54. Academic Press, San Diego, pp. 843–848. https://doi.org/10.1016/B978-0-12- Shiferaw, B., Prasanna, B.M., Hellin, J., Bänziger, M., 2011. Crops that feed the world 6. 416641-7.00096-1. Past successes and future challenges to the role played by maize in global food Core Team, R., 2019. R: A language and environment for statistical computing security. Food Sect. 3 307. https://doi.org/10.1007/s12571-011-0140-5. [Computer software manual]. Vienna,, Austria. Sinche, M., Kannan, B., Paudel, D., Corsato, C., Lopez, Y., Wang, J., Altpeter, F., 2021. Dos-Santos, C.M., Nascimento, W.B., do Nascimento, B.P., Schwab, S., Baldani, J.I., Development and characterization of a Napier grass (Cenchrus purpureus Schumach) Vidal, M.S., 2021. Temporal assessment of root and shoot colonization of elephant mapping population for flowering-time- and biomass-related traits reveal individuals grass (Pennisetum purpureum Schum.) host seedlings by Gluconacetobacter with exceptional potential and hybrid vigor. GCB Bioenergy 13, 1561–1575. https:// diazotrophicus strain LP343. Microbiol. Res. 244 (126651). doi.org/10.1111/gcbb.12876. Fan, C.-C., Su, C.-F., 2008. Role of roots in the shear strength of root-reinforced soils with Suleiman, M.K., Bhat, N.R., Jacob, S., Al-Burais, M., 2018. Performance evaluation of high moisture content. Ecol. Eng. 33, 157–166. Chrysopogon zizanoides under urban conditions of Kuwait. Saudi J. Biol. Sci. 25, Ferreira, O.J.M., Holanda, F.S.R., Pedrotti, A., Vidal Santos, L.D., Silva-Mann, R., 2022. 305–312. https://doi.org/10.1016/j.sjbs.2017.09.012. Root System of Jatropha curcas Provides Resistance and Strength to the Soil. Truong, P., Baker, D., 1998. Vetiver Grass Syst. Environ. Prot. Commun. Soil Sci. Plant Anal. 53, 2955–2967. https://doi.org/10.1080/ Walkley, A., Black, I.A., 1934. An examination of the Degtjareff method for determining 00103624.2022.2099554. soil organic matter, and a proposed modification of the chromic acid titration Foresta, V., Capobianco, V., Cascini, L., 2020. Influence of grass roots on shear strength method. Soil Sci. 37, 29–38. of pyroclastic soils. Can. Geotech. J. 57, 1320–1334. Wang, G., Huang, Y., Li, R., Chang, J., Fu, J., 2020. Influence of Vetiver Root System on Forster, M., Ugarte, C., Lamandé, M., Faucon, M.-P., 2022. Root traits of crop species Mechanical Performance of Expansive Soil: Experimental Studies. Adv. Civ. Eng. contributing to soil shear strength. Geoderma 409, 115642. https://doi.org/ 2020, e2027172 https://doi.org/10.1155/2020/2027172. 10.1016/j.geoderma.2021.115642. Weaich, K., Cass, A., Bristow, K.L., 1992. Use of a penetration resistance characteristic to Goss, M.J., 1991. Consequences of the activity of roots on soil. Plant Root Growth.: Ecol. predict soil strength development during drying. Soil.Tillage Res., Slaking Perspect. 171–186. Hardsetting Soil.: Some Res. Manag. Asp. 25, 149–166. https://doi.org/10.1016/ Gyssels, G., Poesen, J., Bochet, E., Li, Y., 2005. Impact of plant roots on the resistance of 0167-1987(92)90108-N. soils to erosion by water: a review. Prog. Phys. Geogr. 29, 189–217. Whalley, W.R., Watts, C.W., Gregory, A.S., Mooney, S.J., Clark, L.J., Whitmore, A.P., Hall, H.E., Raper, R.L., 2005. Development and concept evaluation of an on-the-go soil 2008. The effect of soil strength on the yield of wheat. Plant Soil 306, 237–247. strength measurement system. Trans. ASAE 48, 469–477. Willatt, S.T., Sulistyaningsih, N., 1990. Effect of plant roots on soil strength. Soil Tillage Hendry, M.T., Barbour, S.L., Martin, C.D., 2014. Evaluating the effect of fiber Res. 16, 329–336. reinforcement on the anisotropic undrained stiffness and strength of peat. J. Geotech. Geoenviron. Eng. 140, 04014054. 11 F. Kumi et al. S o i l & T i ll a g e R e s e a r c h 233 (2023) 105811 Yoon, P.K., 1991. Extracts from a Look-see at Vetiver grass in Malaysia-First progress Zhao, Y., Duan, X., Han, J., Yang, K., Xue, Y., 2018. The main influencing factors of soil report. Vetiver Newsl. 6, 86–96. mechanical characteristics of the gravity erosion environment in the dry-hot valley Zhang, X.Y., Cruse, R.M., Sui, Y.Y., Jhao, Z., 2006. Soil Compaction Induced by Small of Jinsha river. Open Chem. 16, 796–809. Tractor Traffic in Northeast China. Soil Sci. Soc. Am. J. 70, 613–619. https://doi. org/10.2136/sssaj2005.0121. 12