Genesis and Classification of Termite-Mediated Soils along Toposequences in a Semiarid Area of Southeast Ethiopia

Bekele, Abinet; Beyene, Sheleme; Kiflu, Alemayehu; Yimer, Fantaw

doi:https://doi.org/10.1155/2023/7150907

Applied and Environmental Soil Science

On this page

Abstract Introduction Materials and Methods Results and Discussion Conclusions Data Availability Conflicts of Interest Acknowledgments References Copyright Related Articles

Research Article | Open Access

Volume 2023 | Article ID 7150907 | https://doi.org/10.1155/2023/7150907

Genesis and Classification of Termite-Mediated Soils along Toposequences in a Semiarid Area of Southeast Ethiopia

Abinet Bekele,¹Sheleme Beyene,¹Alemayehu Kiflu,¹and Fantaw Yimer²

Academic Editor: M. Vasanthavigar

Received14 Nov 2022

Revised22 Dec 2022

Accepted11 Jan 2023

Published27 Jan 2023

Abstract

Despite the ecosystem functioning they provide, termite pedoturbation along toposequence is overlooked in the genesis of semiarid soils. Therefore, we aimed to describe morphological and physicochemical properties that lead to the classification of termite-mediated soils. In this study, representative pedons, one on each slope class, were described and classified for five different topographical positions, and the soil properties of genetic horizons were compared to those obtained from respective mounds. The result showed that the soils were heavily manipulated by termites except for the pedon at the toe slope. Cambisols were formed on the summit and back slope and resulted from slow pedogenic processes. Luvisols on the toe slope showed redoximorphic features, and gleization and clay synthesis formed the soil, while the upward movement of coarse particles enhanced textural differentiation. Luvisols at the foot of the slope are formed by the partial destruction of iron-bearing minerals accompanied by eluviation-illuviation processes. Accumulation of calcium carbonate following calcification formed Calcisols on the bottom slope. Comparing the mounds and reference pedons, much of the mound’s soil is mined from the subsoil, usually from B horizons. However, their influence on soil properties depended mainly on the topography. Moreover, the morphological and physicochemical properties of the studied pedons exhibited various degrees of variation along topography and clearly showed topographic effects. In conclusion, termites can be a potent mediator of soil genesis across toposequences, and their activities should be considered in the classification and management of semiarid soils. A further retrospective examination of micromorphological evidence is recommended to support this finding.

1. Introduction

Mound-forming termites and large epigeal mounds are the common features of the landscape in most semiarid regions [1, 2]. Termites through their pedoturbation drastically change soil properties and influence soil genesis [3, 4]. The mechanism by which termites affect soil formation is by altering the type of soil that had formed years ago, e.g., from Alisols and Acrisols to umbric properties [4], by extracting clay-enriched materials from the subsoil, e.g., from the Kandic horizon [5] and the Argic horizon [4], by promoting soil homogenization [3], by influencing soil formation processes [6], and by influencing morphological properties [7]. Another important impact is the tunneling and refilling of the subsoil with topsoil [8] and by modifying the mineralogical composition of the soil [9].

Topography is also important in explaining soil variability and pathways of soil transformation. According to Breuning-Madsen et al. [10], macrofauna-induced continuous surface exposure on sloppy landscapes together with episodic precipitation can cause significant downward soil transport. Thuyne and Verrecchia [11] showed that the termite-induced sediment accumulation rate reached up to 1 mm·ha⁻¹·y⁻¹. In such a landscape, profile development can be much more complex than vertically up or down processes since mixing by soil fauna interacts with lateral processes [12].

Although the role of termites in soil formation has been recognized for a long time, only a few studies have been conducted on the pedogenesis of termite-affected soils [3, 4, 13, 14]. In most of these studies, the diversity of pedogenetic processes involved in the genesis of termite-mediated soils is neglected. In Ethiopia, where mound-forming termites predominantly inhibit arid to semiarid ecosystems, termite-mediated soils have been characterized for a specific location [15], and there are still limited studies on the soil genesis of termite-affected soils. Regardless of the differences in soils across topography and the tendency of termites to selectively mine fine particles at different depths [16], the extent to which epigeal termite mounds resemble the reference soils across toposequence is largely unknown. On the other hand, the study of soil characterization and classification that focuses on moisture-deficient areas is limited, and to our knowledge, only the study conducted by Ahmed et al. [17] had initiated in the past.

Since soil is central in an ecosystem, having insight into termite-mediated soils makes this study uniquely contribute to the knowledge of semiarid soils. Additionally, understanding soil-forming processes could also contribute to and have a genetic signal for better soil characterization [18]. Therefore, this study was initiated to characterize the morphological and physicochemical properties of termite-mediated soils, quantify the properties of termite mounds, and compare them with those obtained from genetic horizons which were not influenced by termites.

2. Materials and Methods

2.1. Description of the Study Area

Epigeal termite mounds are the common features of the landscape in a semiarid area of southeast Ethiopia, starting from Dugda Dawa in the north along the highway to Moyale in the south. Geographically, this area lies between 3°53′ to 5°22′ N and 38°16′ to 38°33′ E. Two districts, Dugda Dawa in the north and Miyo in the south, were deliberately selected for this study because the epigeal mounds in the two areas have different physical characteristics. Both study areas are characterized by bimodal rainfall, with the main rainy season occurring between March and May and the short rainy season occurring between October and November. The mean annual rainfall of the Dugda Dawa district is about 711 mm with a mean annual temperature of 19.9°C ranging from 13.3°C to 26.6°C. The rainfall in the Miyo area is 345.5 mm, and the mean annual temperature is 23.9°C, ranging from 18.4°C to 29.5°C. The geology of the parent material is dominated by quartzo-feldspathic gneisses, gneiss and granulite, magnetite-quartz-feldspar gneisses, quaternary metamorphic rock, and sedimentary deposits [19]. The geographic location and the geological conditions of the study area are shown in Figure 1.

2.2. Soil Pedon Description and Sampling

Representative toposequences with widespread termite mounds were selected at the Ilala-Sara and Silala sites in Dugda Dawa and Miyo districts, respectively. The toposequence in Ilala-Sara was grouped into the summit, back slope, and toe slope, while the toposequence in Silala included the foot slope and bottom slope. A pedon was excavated at each topographical position and described in situ following the Guidelines for Soil Description [20], and soil color was determined based on Munsell Soil Color Charts [21]. The soils were classified according to [18]. The location of each pedon was georeferenced using a portable GPS (62-GARMIN).

Additionally, abundance and other termite mound attributes were estimated per topographical position, and termite specimens were collected for taxonomic identification. At each position, three active representative mounds were selected at a distance of 10 m from the reference pedon and considered for soil sampling. A composite soil sample was collected per mound after the entire epigeal mound had been broken up and thoroughly mixed, and a total of three composites were made for termite mounds within a topographical position. However, soil samples were not collected from mounds at toe slopes, as these mounds were less abundant and inactive. All composite soil samples from genetic horizons and mounds were separately tagged and transported to the laboratory.

2.3. Soil Sample Preparation and Analyses

All analyses were performed for the fine Earth fractions (<2 mm). Soil particle size was analyzed by the Bouyoucos hydrometer method [22]. Soil pH and electrical conductivity (EC) were measured in a 1 : 2.5 soil-to-water suspension. Total organic carbon (OC) was determined by wet oxidation in acidified dichromate [23], while available phosphorus was determined using the Olsen method [24]. Exchangeable cations (K⁺, Na⁺, Ca²⁺, and Mg²⁺) and cation exchangeable capacity (CEC) were extracted using 1 M NH₄OAc at pH 7.0 [25]. While Ca²⁺ and Mg²⁺ in the leachate were measured with an atomic absorption spectrophotometer (AAS), K⁺ and Na⁺ were determined with a flame photometer. Calcium carbonate was estimated by HCl neutralization and titration with NaOH [26] and reported as calcium carbonate equivalent (CCE). Available micronutrients (Zn, Mn, Fe, and Cu) were extracted using the diethylenetriaminepentaacetic acid method, and the contents in the leachate were measured using AAS.

3. Results and Discussion

3.1. Sites and Morphological Characteristics of Termite Mounds

The site characteristics of the pedons showed variations in drainage, degree of erosion, and surface stoniness (Table 1). Pedons D01, D02, and D03 are located on the summit, back slope, and toe slope positions, respectively, while pedon M01 is located on the foot slope and pedon M02 is found on the bottom slope. Pedons D01 and D02 were formed from saprolitic materials, while pedon D03 was developed from alluvium deposits set back from slope sediments. Pedon M01 was derived from finely textured eluvium residuum, while pedon M02 was established from lacustrine deposits. Although termite mounds inhibited the survey sites, pedon M01 indicated intense termite activity, as sheetings were formed overnight in an open pit.

Two termite genera representing the studied toposequences were identified. The genus Macrotermes built open, cathedral-shaped mounds that reached up to 4 m high at all slope positions, except at the bottom slope, where it formed closed conical mounds with a mean height of up to 3.2 m (Table 2). The other termite genus, Odontotermes, built closed, dome-shaped mounds at the toe slope with a mean height of up to 2.2 m. Both genera are fungus-growing termites that are commonly reported in Africa [27] as well as the study area [1, 28]. These genera influence savanna soils much more than the other termite groups [11]. The abundance of Macrotermes mounds reached up to 23 mounds ha⁻¹, while the Odontotermes genera had about 4 mounds ha⁻¹. Macrotermes deposit up to 12.51 m³·ha⁻¹, demonstrating an important role for termites in soil pedogenesis. According to Wilkinson et al. [29], abundant and massive mounds on the summit and back slopes may be related to saprolite because saprolites buffer water availability during climate fluctuations for soil organisms [30]. In general, mounds on gentle slopes had a larger structure than mounds on a nearly flat landscape. Larger mounds have stable soil chemistry [31]. Accordingly, termite pedoturbation is intense in the gently undulating landscapes of the study area. In support of this result, Abe et al. [6] reported that the influence of termites on pedogenesis is more evident in inclined than level positions.

3.2. Differentiation of Soil Morphology across the Toposequences

The soils of the study area were characterized by deep pedons, typically over 200 cm (Table 3). The thinnest A horizon was observed in pedon D02, while the thickest surface horizon was in pedon M01. Pedons with deeper depths may be due to the nature of the parent materials, which favor permeability to the depth at which termites search for fine particles. Thick surface horizons could be acquired due to erosion off the mounds and galleries which redistributed over the landscape and resulted in deeper soil depth [32]. In contrast, thin horizons could be related to the removal of surface materials as a result of topographic effects.

Horizontal boundaries varied along the toposequences, and pedons D01 and D02 had diffuse and gradual wavy boundaries at the surface and gradual wavy boundaries at the subsurface (Table 3). While gradual wavy boundaries may be associated with underlying saprolitic materials, wavy horizon topography likely accentuates the presence of lamellae [33]. According to FAO [34], biological pedoturbation leads to diffuse or gradual horizon boundaries. A pedon D03 had a clear wavy horizon boundary superimposed on an abrupt smooth horizon, reflecting the mode of deposition [35]. Pedons M01 and M02 had clear to gradual smooth horizon boundaries, which were probably facilitated by termite pedoturbation along with fine to medium grassroots [36, 37]. In general, gradual distinctness in most pedons is attributed to the systematic and continuous supply of soil materials [38], whereas wavy and smooth boundaries signify the removal of soil materials due to topography and reworked soil horizons [39].

Pedons vary greatly in color, and soil redness decreases down the slope (Table 3). Consistent with the findings by Jimoh et al. [40], small amounts of OC mixed with soil particles resulted in a reddish-brown surface soil color across the toposequences that becomes lighter with decreasing OC in the subsurface of the pedons. In most pedons, the hue pattern began to emerge by 2.5YR, although there were some irregularities. The consistent hue pattern could be due to a termite-induced translocation of soil materials. All examined pedons had a low value and high chroma, except for pedon M02, which tended to show a high value and low chroma. Conspicuously, low value and high chroma without apparent iron illuviation in the pedons D01, D02, and M01 may be related to illuvial sesquioxide accumulation. Due to the proximity of pedon D03 to seasonal floods, the reddish-brown color of the surface horizon might be owing to oxidized iron, while a very dark gray subsurface horizon indicates iron reduction. This result is consistent with the findings of Romanens et al. [41] who observed Fe coatings that give a reddish color to the top horizons without any contact with underlying horizons. Regarding the soil color of termite mounds, representative mounds were red (2.5 YR 4/6, dry) on the summit and back slope, dark red (2.5 YR 3/6, dry) on the foot slope, and white (5 YR 8/1, dry) on the bottom slope. The similarity of termite mound color to subsurface horizons in reference pedons supports the fact that termites collect subsurface soil particles for mound formation [3, 7, 15].

The surface soil of the studied pedons had a weak fine to medium granular, crumbly, and subangular blocky texture that changed to a weak medium weak fine to very coarse angular blocky texture, a weak medium subangular blocky or prismatic texture, and very strong and very coarse rock in the B horizons and massive rocks in the C horizons (Table 3). The structural unit of the topsoil may be related to termite pedoturbation. This is consistent with the recent findings of De Freitas et al. [3], which indicated that soil that has been worked a lot by termites exhibited granular and small subangular block structures. According to Bera et al. [7], subangular structures with a reddish color predominated also in the subsoil, as pedoturbation decreased with depth. In accordance with Dinssa and Elias [42], the angular block structure formed in the subsoil of D02 could be due to an overlying dense soil layer, while that of D03 and M01 might be related to enriched clay particles, reduced OM, and root distribution.

The soil consistency was less differentiated among the topographical positions (Table 3). With some exceptions, all pedons are dominated by soft, slightly hard to hard dry consistencies, with moist consistencies ranging from loose, very friable to extremely firm. These pedons were generally marked by a slight stickiness ranging from nonplastic to plastic. The granular or fine subangular blocky structures probably favored the friable and very friable consistencies at the surface horizons of the pedons D01, M01, and M02 due to moderate to high OC levels. The representation of wet consistence in the subsoil of most horizons is attributed to reduced OM and clay accumulations [42]. A soft, dry consistency in the surface horizon of M02 may underpin secondary carbonate. The lack of stickiness and plasticity in some horizons could be related to the amount of clay particles in the soil. In contrast, that of pedon M02 could be attributed to indurate carbonate.

The surface layers of pedons D01 and D02 had very few fine gravels (>2 mm), while gravelly horizons were exposed to the surface in pedon D02 (Table 3). The subsurface horizons of these pedons were marked by abundant coarse gravels and stones that changed to very few fine gravels with depth. Pedon D03 had very few medium gravels near the surface layer that decreased to zero with depth. In pedon M02, rock fragments were distinguished in most of the subsurface horizons, and these exceeded more than 80% in the lowest horizons. Gravelly horizons in pedons D01 and D02 are probably due to fragments caused by dominant physical weathering and are not related to termite stone lines. This is because a termite-related stone layer is identified by the distinguished chemical properties of the soil overlying the layer [14]. In these pedons, a reduced fragment distribution with depth indicates downslope transported soil materials [33]. The fragments are exposed to the surface layers on the back slope indicating the influence of topography since surface erosion is higher than the rate of replenishment.

3.3. Physical Properties of Termite-Mediated Soils

Both soil particle sizes (≤2 mm) and silt: clay ratios are irregularly distributed over topographical positions (Table 4). The absence of a definite pattern of soil particle sizes is probably attributed to surface soil erosion and subsequent accumulation which may influence pedogenic processes [43]. Sand particles were gradually decreased in all pedons, except for pedon M02 which exhibited an increasing pattern with depth. In line with these results, a gradually decreasing sand pattern was also reported by Bonifacio et al. [44]. According to Jimoh et al. [40], the trends of sand distribution in pedon M02 might be attributed to irregular deposition. Based on the rating by Hazelton and Murphy [45], all pedons had a high to very high sand content, while the silt content was classified as very low to low. The high to very high sand content could be related to the influence of parent materials. Given that the soils are in the moisture deficit region, such low silt contents were common. According to Daher et al. [46], negligible ornithogenesis and sulphides probably impede the advanced weathering of silts and clays in semiarid regions.

A steady increase in clay content was evident in the pedons D01, D02, and M01, showing accumulation with depth. The peak clay content of pedon M01 could be related to clay illuviation, while that of pedons D01 and D02 might have been inherited from the saprolite materials, which also resulted in a low silt-to-clay ratio (Table 4). This could be attributed to the fact that saprolites are more clay-rich because the particles are trapped before seeping into the deeper layer [47]. Clay particles varied rapidly with depth in the pedon D03, and such a wetting-front pattern could be associated with heavy illuviation [48]. Based on the CEC of clay, however, the absence of 2 : 1 clay in the upslope soil (pedons D01 and D02) implies that clay accumulation in pedon D03 is not of colluvial origin, whereas the irregular clay pattern within this pedon would not support in situ clay formation [49]. Based on these views, it appears that clay synthesis in the pedon D03 comes from alluvial deposits. On the other hand, clay content showed minimum and maximum peaks with depth in the pedon M02. According to Minasny et al. [48], soil attributes can show a minimal depth pattern when surface mixing is excessive and a translocation process is operating. In this pedon, consistent texture throughout the depth might be inherited from carbonate-rich parent materials [41].

The average clay content of the mounds at the summit, back slope, and foot slope was reduced by 14%, 112%, and 44% compared to the maximum content in pedons D01, D02, and M01, respectively. While the clay content of the mounds was 50% higher than the maximum clay content in the subsurface of pedon M02. In this study, termite pedoturbation affected soil particle-sizes by enriching mounds with silt and clay compared to reference pedons, but the magnitude was related to the topographical positions. Consistent with this study, Freitas et al. [3] reported that a wide range of particles ranging from silt to clay could be selected for mound building. Clay enrichment in termite mounds is related to translocated mica and smectite or saprolitic materials from the subsoil [50]. According to Jouquet et al. [5], high clay content in termite mounds could arise from the dissolution of plant matter, while resilication is also another process, particularly for smectite.

Except at the bottom slope, the studied mounds were predominantly sandy clay loam in texture, and termite mound soils were not different from matched reference pedons, corroborating the findings of Freitas et al. [3]. According to Abe and Wakatsuki [49], the clay mineralogy of the termite mound soil is also very similar to adjacent bulk soils. A contrasting mound soil texture, as with the pedon D02, implies that termites mine and translocate soil particles depending on the nature of the subsoil. In addition, the silt-to-clay ratio of termite mounds ranged from 0.37 to 0.61 which was comparable to the ratio obtained from reference pedons (Table 4). This supports previous studies [50, 51] and indicates that termites relocate fine particles from the subsurface of the respective reference soils across toposequences.

3.4. Chemical Properties of Termite-Mediated Soils

Soil pH (1 : 2.5 H₂O) varied across the toposequences (Table 4), which could be attributed to the leaching of cations from the upslope and the subsequent accumulation at the nearly level positions [52]. Moderately acidic pH increased with depth in pedon D01 and became strongly acidic at the subsurface of pedon D02, while it was uniformly distributed in pedon D03. The soil pH was neutral at the surface, slightly increased with depth in pedon M01, and changed to moderately alkaline in pedon M02. Apart from pedon M02, the acidic soil surface could be related to the leaching of basic cations, while limited leaching due to herbaceous vegetation probably maintained a near neutral soil reaction in pedon M01, whereas the alkalinity of M02 could be attributed to the nature of the parent material. In general, the comparable soil pH within the pedons likely indicates slow weathering and soil development. The soil pH values of termite mounds ranged from slightly acidic to neutral which were not similar to the respective pedons D01 and D02 but similar to the subsurface horizons of the pedons M01 and M02 (Table 4). The soil pH of mounds depends on the corresponding pH values of the reference soil at a given topographical position. This is consistent with the findings of Sarcinelli et al. [32]. According to Bera et al. [7], the pH of termite mounds is attributed to cation-clay accumulation and the rate of OM decomposition in the mounds.

While no carbonate was present in all other soils, the soils in the pedon M02 invariably exhibited strong effervescence with dilute HCl (10%) and showed calcium carbonate that rapidly increased with depth (Table 4). Termite mounds at the same position had CCE that was 2.5 times higher than the respective surface soil and comparable to the subsurface of the reference pedon. The accumulation of carbonate within a pedon could be inherited from parent materials and other associated rocks. The CCE-depth relationship also supports the premise that it is geological rather than pedogenic since weathering of the bedrock results in substantial Ca²⁺ [41]. According to Buss et al. [53], the carbonate incidence near the surface soil suggests a limited percolation of water to leach carbonate. The distinct features of CCE in mounds indicate that termite pedoturbation concentrates carbonate and relays it to bulk soils. This result is consistent with the observation of Mcauliffe et al. [54] who reported the enrichment of mounds with carbonate.

Soil EC content showed an irregular pattern across topographical positions owing to leaching and accumulation of basic cations (Table 4). Apart from pedon M02, the EC in the horizons of the pedons was below 2.34 mS cm⁻¹. For the pedon M02, the EC value of 8.66 mS cm⁻¹ in the surface soil decreased gradually with depth. In general, negligible EC differences in the soils of most pedons indicate limited eluviation and illuviation processes, while the amounts in the pedon M02 could be related to the substantial basic cations in the soils of the pedon. Compared to the reference pedon, the average EC of mounds was almost seven times higher than the maximum value for soils in the pedons D01 and M01 and 43 times higher in the pedon D02 while comparable in the pedon M02. By and large, the higher EC of termite mounds compared to bulk soils could be attributed to more soluble ions in the biogenic structure [55].

The soil OC varied consistently along the toposequences (Table 4). This corroborates with the findings of Sarcinelli et al. [32] which suggest that similar compositions of termite foraging material result in comparable OC content across topographical positions. Except for the pedon D03 which had a wetting-front pattern, the OC gradually decreased with depth. The OC of mounds at the summit and back slope exceeded the higher content of the surface horizons by more than 70%. However, the OC content of the mounds was lower than that of the surface bulk soils of the pedons M01 and M02. While wetting-front OC trends might be attributed to slow translocation, the gradual decreasing pattern probably indicates the accumulation of OC through grass decay. Based on the OC-to-clay index stated by Merante et al. [56], the high to moderate OC content in pedon M02 could also be associated with a complex OC with clay content, which could lead to C stabilization. Regarding the OC levels of the mounds, the result is consistent with the observation of Obi et al. [57] and demonstrates topographic effects. In addition to the high mineralization rate in a semiarid area, the fact that fungus-growing termites mineralize about 20% of the total OC [11] probably results in a low OC content.

The available P was less than 10.69 mg·kg⁻¹ across the studied pedons and irregularly distributed across toposequences (Table 4). The distribution pattern of P showed accumulation with depth in the pedons D01, D03, and M02, while its value gradually decreased in the pedons D02 and M01. Compared to the reference pedon, the average P content of the mounds was higher by 51%, 163%, and 78% than the maximum values of the soil horizons in the pedons D02, M01, and M02, respectively. In general, low available P could be attributed to insoluble Ca phosphates comprising the calcareous materials in the pedon M02, while the low levels in the other pedons might be due to sorption by sesquioxides. According to Dinssa and Elias [42], a decreasing P-trend with depth could be associated with reduced OM and P fixation. The highly available P of termite mounds could be related to the incorporation of OM into mounds [4]. However, Bera et al. [7] reported fewer contents in mounds arguing that P content depends on the termite feeding habit and the soil materials. The P content of the mound also depends on the amount of clay, pH, types, and termite species involved in mound formations [11]. In our case, the viable explanation for the highly available P content of mounds is probably related to the soil pH nearing neutral reaction, clay content, and long-term accumulation of OM leading to humification which in turn favors the release of P.

The CEC of the soils varied irregularly across the toposequences (Table 4). In pedons D01 and D02, CEC had an irregular pattern with depth. With a rapidly changing pattern, the pedons D03 and M01 had increasing CEC values with depth, while pedon D03 had substantially high CEC in the lowest horizon. The CEC gradually decreased with depth and accumulated in the lowest horizon of the pedon M02. A low CEC in the surface soil could be associated with a low OM content, while a moderate to high CEC in the subsurface could result from clay accumulation. A distinctly high CEC in the pedon D03 could be due to clay synthesis (r > 0.99, not shown here) and the accumulation of Ca²⁺ and Mg²⁺ (r = 0.99). On the other hand, the considerable CEC in the pedon M02 could be associated with substantial clay and Ca²⁺ in the soils [41]. The CEC values of the mounds on the summit and back slope were three times, while that of the mounds at the foot slope and bottom slope were two-fold higher than the maximum CEC in the respective horizons of the reference pedons. The higher CEC of termite mounds might be independent of OM (r = 0.19 ns), and it is rather related to the presence of 2 : 1 clay. This is because the CEC of both soil and clay is higher than what would be expected for soils dominated by kaolinite, which suggests the presence of montmorillonite clay in the soil of termite mounds and related to the subsoil of respective pedons. This corroborates with the findings of Abe and Wakatsuki [49] who observed minor modification of clay mineralogy in Macrotermes mounds compared to their respective soils. Consequently, this clay probably explains the high CEC of termite mounds.

While Ca²⁺, Mg²⁺, and K⁺ showed an increasing trend, Na⁺ exhibited decreasing patterns across topographical positions (Table 5). The increasing pattern of Ca²⁺ and Mg²⁺ down the slope could be due to lateral movement. Exchangeable Ca²⁺ was less than 10 cmol_c kg⁻¹ in all pedons except for pedon M02 and a distinctly high value in the lowermost horizon of pedon D03. Both Ca²⁺ and Mg²⁺ had an irregular pattern with depth in the pedons D01 and D02 and M02, rapidly changed with depth in the pedon D03, and showed accumulation in the subsurface of the pedon M01. The high Ca²⁺ in the pedon D03 could be related to the CaCO₃ accumulation [53] and that of the pedon M02 could be attributed to the weathering of carbonate-enriched parent materials [41]. Exchangeable K⁺ gradually increased in the pedons D01 and D02, while the other pedons gradually decreased with depth. In all pedons, Na⁺ had an irregular decreasing trend with depth except within pedon D03 which had some enrichment. The increase of K⁺ with depth in the pedons D01 and D02 could be related to the leaching of cations [40]. Relatively, the high K⁺ in the pedon D03 is probably due to the accumulation of mica at that position, while the Na⁺ accumulation in the lowest horizon of this pedon may be due to its greater mobility. The Ca > Mg > K > Na order in the studied pedons might be due to the release of these ions by weathering before the rocks form. The major cations that dominated the exchange complex of the pedons are consistent with the findings by Beyene [58].

The mounds at the summit and back slope had 4 to 10 times higher Ca²⁺, while the mounds at the bottom slope had 57 times higher Ca²⁺ than the maximum value in the respective reference pedon (Table 5). On the other hand, the Mg²⁺ contents of the mounds were lower compared to that of the surface soils but comparable to the levels in the subsurface horizon of the reference pedons. The K⁺ content of mounds decreased up to 52% compared to the maximum values in the soils of the reference pedons D01 and D02, while the K⁺ content of the mounds was higher by more than 28% in the pedons M01 and M02. The exchangeable Na⁺ of the mounds was decreased by 7% to 59% compared to the maximum values in the reference pedons. While termite pedoturbation did not affect the Na⁺ and Mg²⁺ across the toposequences, their effects were evident for Ca²⁺ at all locations and K⁺ at the foot slope and bottom slope positions. In support of this study, Obi et al. [57] also considered Ca²⁺, Mg²⁺, and K⁺ as indicators of pedoturbation.

3.5. Available Micronutrients

While Zn and Cu decreased down the slope, Fe and Mn showed irregular patterns across toposequences (Table 5). Consistent with Hailu et al. [43], the patterns of Fe and Mn were similar at any given topographical position because of their similar chemical behavior. The micronutrient contents were in the order of Fe > Mn > Zn > Cu. According to Zhu et al. [59], the dominance of Fe at any given topographical position might be due to the dissolution of metal oxide in the clay particles which in turn strongly depends on the slope position while the availability of Zn and Cu related to desorption from clay, carbonate, and OM surface. Available Fe, Mn, and Zn contents showed an irregular pattern with depth in the pedons D01, D02, and M02, probably owing to termite pedoturbation while accumulating with depth in pedon D03. While Fe and Mn accumulated just below the surface layer, Zn gradually decreased with depth in the pedon M01. Available Cu showed a gradual decrease with depth in the pedons D01, D02, and M02, while it gradually increased in the pedons D03 and M01.

3.6. Pedogenesis and Classification of the Soils

The subsurface horizon of the pedons D01 and D02 showed an incipient pedogenetic alteration, with half of the original rock structure being obliterated. Additionally, the horizon thickness and texture of the soils qualify them for a cambic horizon. The cambic horizons were also specifically identified by redder hues (2.5 years) and higher chroma (>6) on the surface than the underlying horizons. On the other hand, sufficiently intense reduction and segregation of iron which resulted in low chroma were indicative of the identification of the cambic horizon in the pedon D03. Prominent clay accumulation in the subsoil horizons of the pedon D03 qualified argic diagnostic horizon. The pedon M01 was differentiated by a redder hue, higher chroma, and an accumulation of clay and sesquioxides. The moist color changes from reddish-brown in the upper to dark reddish-brown at the subsurface horizon substantiating clay illuviation for identification of the argic horizon. Termite pedoturbation [13], shrubs, perennials, and grasses probably resulted in clay accumulation in this pedon. A polyhedral structure in this pedon did not meet the diagnostic features of a nitic horizon. The subsurface horizon of the pedon M02 had more than 15% CCE in the fine Earth fraction which meets the diagnostic criteria for a calcic horizon. According to Macedo et al. [60], the absence of secondary carbonate accumulation may be due to termite pedoturbation. Based on the above diagnostic horizons and using established criteria from IUSS Working Group WRB [18], the pedons D01 and D02 were classified as Cambisols. Argic horizons accompanied by CECs of clay greater than 24 cmol_c kg⁻¹ in the pedons D03 and M01 differentiated the soils into Luvisols, whereas the pedon M02 was identified as Calcisols. The principal and supplementary qualifiers are summarized in Table 6.

3.7. Pedogenic Processes

The soils in the surface layers of all pedons showed the feature of termite pedoturbation as traced by granular or subangular blocky structures and gradual or diffuse horizon boundaries except for soils in pedon D03. In addition, incipient pedogenic processes, as shown by little alteration and very few fine root distributions, might be some of the soil-forming processes in the pedons D01 and D02. As stated by Borden et al. [52], the loss of cations through downslope translocation also probably promotes desilication in these two pedons. The evidence of gleization was observed in the pedon D03, but cyclic drying and wetting in the surface and the remains of moisture in the subsoil favored hydrolysis which subsequently derived clay formation. In this pedon, the absence of mottling suggests minimal current pedogenetic activities other than gleying and illuviation processes. According to Bonifacio et al. [44], gleyic properties indicate strong weathering of the iron-bearing minerals probably as evidenced by the depletion of available Fe on the surface of the pedon. In the pedon M01, desilication, braunification, rubification, ferrugation, leucinization, and illuviation might be the most important pedogenic processes. Calcification was the dominant process in the pedon M02 while humification dispersed to a certain extent throughout the solum. Following Deressa et al. [36], there was a textural differentiation in examined pedons except for the pedon M02 which showed a clay contrast index of 0.87 indicating a lower textural gradient. In addition to termite pedoturbation, calcification processes result in homogenous clay distribution within the pedon M02.

4. Conclusions

Termite pedoturbation promotes soil genesis by modifying soil morphology, sorting particles, and altering physicochemical properties. However, the magnitude of their impact is, by and large, specific to topographical positions. The studied soil properties also show topographic effects exhibited by various degrees of variation along the toposequences where gently sloped positions had the highest clay, K⁺, and Na⁺, while sand, pH, OC, Ca²⁺, and Mg²⁺ were high in nearly level positions. This study also showed the reference subsoils are mined for fine particles, and thus, there is a slight modification to B horizons across the toposequences. The soils of the study sites are grouped into Cambisols, Luvisols, and Calcisols. Pedogenic processes that formed Cambisols on the summit and back slope are less active processes complemented by termite pedoturbation. Luvisols are formed by gleying and clay synthesis at the toe slope where termites are absent or by illuviation processes and termite pedoturbation on the foot slope. Calcisols occur to a lesser extent on nearly level land, while calcification and termite pedoturbation mainly operated during pedogenesis. Based on the results of this study, termite pedoturbation and topography should be carefully considered during the classification and management of semiarid soils. Although this study marked the influence of termite activity and topography, the precise effects are warranted following micropedological evidence.

Data Availability

The data used to support the findings of this study are available upon request.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

The authors acknowledge Ferdu Azerefegne (PhD) for his contribution to termite identification. The authors also thank the Soil and Plant Nutrition Laboratory and the staff of the lab in the School of Plant and Horticultural Sciences, College of Agriculture, Hawassa University for all the facilities and support they provided. Finally, the authors are also grateful to Bule Hora University for its logistic support throughout the study period. The research work was financially supported by the Ethiopian Ministry of Education in the frame of Ph.D. subsidiary funds.

References

D. G. Debelo, “Faunal survey of the termites of the genus macrotermes (isoptera: termitidae) of Ethiopia,” Journal of Entomology and Nematology, vol. 10, no. 7, pp. 50–64, 2018.
View at: Publisher Site | Google Scholar
D. G. Debelo and E. G. Degaga, “Termite species composition in the central rift valley of Ethiopia,” Agriculture and Biology Journal of North America, vol. 5, no. 3, pp. 123–134, 2014.
View at: Publisher Site | Google Scholar
D. F. De Freitas, J. C. Ker, L. A. da S. Filho, T. T. C. Pereira, O. F. F. de Souza, and C. E. G. R. Schaefer, “Pedogeomorphology and paleoenvironmental implications of large termite mounds at the Brazilian semiarid landscape,” Geomorphology, vol. 387, Article ID 107762, 2021.
View at: Publisher Site | Google Scholar
D. Rückamp, C. Martius, L. Bornemann, D. Kurzatkowski, L. P. Naval, and W. Amelung, “Soil genesis and heterogeneity of phosphorus forms and carbon below mounds inhabited by primary and secondary termites,” Geoderma, vol. 170, pp. 239–250, 2012.
View at: Publisher Site | Google Scholar
P. Jouquet, F. Jamoteau, S. Majumdar et al., “The distribution of Silicon in soil is influenced by termite bioturbation in South Indian forest soils,” Geoderma, vol. 372, Article ID 114362, 2020.
View at: Publisher Site | Google Scholar
S. S. Abe, S. Yamamoto, and T. Wakatsuki, “Physicochemical and morphological properties of termite (Macrotermes bellicosus) mounds and surrounding pedons on a toposequence of an inland valley in the southern Guinea savanna zone of Nigeria,” Soil Science & Plant Nutrition, vol. 55, no. 4, pp. 514–522, 2009.
View at: Publisher Site | Google Scholar
D. Bera, S. Bera, and N. Das Chatterjee, “Termite mound soil properties in West Bengal, India,” Geoderma Reg, vol. 22, Article ID e00293, 2020.
View at: Publisher Site | Google Scholar
J. A. Kristensen, K. J. Thomsen, A. S. Murray, J.-P. Buylaert, M. Jain, and H. Breuning-Madsen, “Quantification of termite bioturbation in a savannah ecosystem: application of OSL dating,” Quaternary Geochronology, vol. 30, pp. 334–341, 2015.
View at: Publisher Site | Google Scholar
J. Van Thuyne, I. Darini, A. Mainga, and E. P. Verrecchia, “Are fungus-growing termites super sediment-sorting insects of subtropical environments,” Journal of Arid Environments, vol. 193, Article ID 104566, 2021.
View at: Publisher Site | Google Scholar
H. Breuning-Madsen, J. A. Kristensen, T. W. Awadzi, and A. S. Murray, “Early cultivation and bioturbation cause high long-term soil erosion rates in tropical forests: OSL based evidence from Ghana,” Catena, vol. 151, pp. 130–136, 2017.
View at: Publisher Site | Google Scholar
J. Van Thuyne and E. P. Verrecchia, “Impacts of fungus-growing termites on surficial geology parameters: a review,” Earth-Science Reviews, vol. 223, Article ID 103862, 2021.
View at: Publisher Site | Google Scholar
P. Lavelle and A. V. Spain, “Soil formation,” in Soil Ecology, pp. 143–202, Kluwer Academic Publishers, Dordrecht, Netherlands, 2003.
View at: Google Scholar
H. J. de Souza, J. H. C. Delabie, and G. A. Sodré, “Termite participation in the soil-forming processes of ‘murundus’ structures in the semi-arid region of Brazil,” Revista Brasileira de Ciência do Solo, vol. 44, Article ID e0190133, 2020.
View at: Publisher Site | Google Scholar
M. Smith, I. Ward, and I. Moffat, “How do we distinguish termite stone lines from artefact horizons? A challenge for geoarchaeology in tropical Australia,” Geoarchaeology, vol. 35, pp. 1–11, 2019.
View at: Publisher Site | Google Scholar
A. Tilahun, C. Yamoah, H. Erens, B. B. Mujinya, A. Verdoodt, and E. V. Ranst, “Quantifying the masses of Macrotermes subhyalinus mounds and evaluating their use as a soil amendment,” Agriculture, Ecosystems & Environment, vol. 157, pp. 54–59, 2012.
View at: Publisher Site | Google Scholar
P. Jouquet, M. Lepage, and B. Velde, “Termite soil preferences and particle selections: strategies related to ecological requirements,” Insectes Soclaux, vol. 49, pp. 1–7, 2002.
View at: Publisher Site | Google Scholar
A. Ahmed, N. Dechassa, S. Gebeyehu, and Y. Alemayehu, “Characterization of soil of jijiga plain in the Somali regional state of Ethiopia,” East African Journal of Sciences, vol. 12, no. 1, pp. 1–10, 2018.
View at: Google Scholar
Iuss Working Group Wrb, “World Reference Base for Soil Resources 2014, update 2015 International soil classification system for naming soils and creating legends for soil maps,” Tech. Rep., FAO, Rome, Italy, 2015, World Soil Resources Reports No. 106.
View at: Google Scholar
B. É. Ó. Dochartaigh, “User guide: Africa groundwater atlas country hydrogeology maps, version 1.0,” British Geological Survey, Nottingham, UK, 2019, Open Report, Or/19/035.
View at: Google Scholar
Fao, in Guidelines for Field Soil Descriptions, Food and Agriculture Organization of the United Nations, Rome, Italy, 4 edition, 2006.
Munsell Color, “Munsell soil color Charts,” Macbeth Division of Kollmorgen Instruments Corporation, New Windsor, NY, USA, 1954.
View at: Google Scholar
G. J. Bouvoucos, “A recalibration of the hydrometer method for making mechanical analysis of soils,” Agronomy Journal, vol. 43, pp. 434–438, 1951.
View at: Google Scholar
D. W. Nelson and L. E. Sommers, “Total carbon, organic carbon, and organic matter,” in Methods of Soil Analysis: Chemical and Microbiological Properties, No 9 (Part 2) in the Series Agronomy, A. L. Page, R. H. Miller, and D. R. Keeney, Eds., pp. 539–577, American Society of Agronomy, Inc. and Soil Science Society of America, Inc, Madison, Wisconsin, USA, 2 edition, 1982.
View at: Google Scholar
S. R. Olsen, C. V. Cole, F. S. Watanabe, and L. A. Dean, Estimation of Available Phosphorus in Soils by Extraction with Sodium Bicarbonate, US Department of Agriculture, Washington, DC, USA, 1954.
G. W. Thomas, “Exchangeable cations,” in Methods of Soil Analysis: Chemical and Microbiological Properties, No 9 (Part 2) in the Series Agronomy, A. L. Page, R. H. Miller, and D. R. Keeney, Eds., pp. 159–164, American Society of Agronomy, Inc. and Soil Science Society of America, Inc, Madison, Wisconsin, USA, 2 edition, 1982.
View at: Google Scholar
Fao, Standard Operating Procedure for Soil Calcium Carbonate Equivalent, Titrimetric method. Food and Agriculture Organization (FAO) of the United Nations Rome, Rome, Italy, 2020.
H. M. O. Otieno, “Impacts and management of termites (isoptera: termitidae) among smallholder farmers in east Africa,” J. Agric. Ecol. Res. Int., , vol. 16, no. 1, pp. 1–12, 2018.
View at: Publisher Site | Google Scholar
L. Menichetti, L. Landi, P. Nannipieri, T. Katterer, H. Kirchmann, and G. Renella, “Chemical properties and biochemical activity of colonized and abandoned litter-feeding termite (macrotermes spp.) mounds in chromic cambisol area on the,” Pedosphere, vol. 24, no. 3, pp. 399–407, 2014.
View at: Publisher Site | Google Scholar
M. T. Wilkinson, P. J. Richards, and G. S. Humphreys, “Breaking ground: pedological, geological, and ecological implications of soil bioturbation,” Earth-Science Reviews, vol. 97, pp. 257–272, 2009.
View at: Publisher Site | Google Scholar
J. C. B. dos Santosa, E. L. Pera, V. S. J. deSouza et al., “Porosity and genesis of clay in gneiss saprolites: the relevance of saprolithology to whole regolith pedology,” Geoderma, vol. 319, pp. 1–13, 2018.
View at: Publisher Site | Google Scholar
J. Muvengwi, H. G. T. Ndagurwa, T. Nyenda, and M. Mbiba, “Nutrient dynamics and plant assemblages of Macrotermes falciger mounds in a savanna ecosystem,” Acta Oecologica, vol. 76, pp. 13–21, 2016.
View at: Publisher Site | Google Scholar
T. S. Sarcinelli, C. Schaefer, L. Lynch et al., “Chemical, physical and micromorphological properties of termite mounds and adjacent soils along a toposequence in Zona da Mata, Minas Gerais State, Brazil,” Catena, vol. 76, pp. 107–113, 2009.
View at: Publisher Site | Google Scholar
Soil Survey staff, Soil Taxonomy a Basic System of Soil Classification for Making and Interpreting Soil Surveys, United States Department of Agriculture Natural Resources Conservation Service Agriculture, Washington, DC, USA, 2 edition, 1999.
Fao, “Lecture notes on the major soils of the world,” Food and Agriculture Organization of the United Nations, Rome, Italy, 2001, World Soil Resources Reports, 94.
View at: Google Scholar
J. Waroszewski, M. Malkiewicz, R. Mazurek, B. Labaz, P. Jezierski, and C. Kabala, “Lithological discontinuities in Podzols developed from sandstone cover beds in the Stolowe Mountains (Poland),” Catena, vol. 126, pp. 11–19, 2015.
View at: Publisher Site | Google Scholar
A. Deressa, M. Yli-halla, M. Mohamed, and L. Wogi, “Soil classification of humid Western Ethiopia: a transect study along a toposequence in Didessa watershed,” Catena, vol. 163, pp. 184–195, 2018.
View at: Publisher Site | Google Scholar
B. Labaz, E. Musztyfaga, J. Waroszewski, A. Bogacz, P. Jezierski, and C. Kabala, “Landscape-related transformation and differentiation of chernozems – catenary approach in the silesian lowland, SW Poland,” Catena, vol. 161, pp. 63–76, 2018.
View at: Publisher Site | Google Scholar
J. B. Kowalska, T. Zaleski, A. Józefowska, and R. Mazurek, “Soil formation on calcium carbonate-rich parent material in the outer Carpathian Mountains–A case study,” Catena, vol. 174, pp. 436–451, 2019.
View at: Publisher Site | Google Scholar
H. Jäger, M. Achermann, J. Waroszewski et al., “Pre-alpine mire sediments as a mirror of erosion, soil formation and landscape evolution during the last 45ka,” Catena, vol. 128, pp. 63–79, 2015.
View at: Publisher Site | Google Scholar
A. I. Jimoh, A. L. Mbaya, D. Akande, D. T. Agaku, and S. Haruna, “Impact of toposequence on soil properties and classification in Zaria, Kaduna State, northern Guinea Savanna, Nigeria,” EQA - International Journal of Environmental Quality, vol. 38, pp. 48–58, 2020.
View at: Publisher Site | Google Scholar
R. Romanens, F. Pellacani, A. Mainga, R. Fynn, P. Vittoz, and E. P. Verrecchia, “Soil diversity and major soil processes in the Kalahari basin, Botswana,” Geoderma Reg, vol. 19, Article ID e00236, 2019.
View at: Publisher Site | Google Scholar
B. Dinssa and E. Elias, “Characterization and classification of soils of bako tibe district, west shewa, Ethiopia,” Heliyon, vol. 7, Article ID e08279, 2021.
View at: Publisher Site | Google Scholar
A. H. Hailu, K. Kibret, and H. Gebrekidan, “Characterization and classification of soils of kabe subwatershed in south wollo zone, northeastern Ethiopia,” African J. Soil Sci., , vol. 3, no. 7, pp. 134–146, 2015.
View at: Google Scholar
E. Bonifacio, G. Falsone, G. Simonov, T. Sokolova, and I. Tolpeshta, “Pedogenic processes and clay transformations in bisequal soils of the Southern Taiga zone,” Geoderma, vol. 149, no. 1–2, pp. 66–75, 2009.
View at: Publisher Site | Google Scholar
P. Hazelton and B. Murphy, Interpreting Soil Test Results: What Do All the Numbers Mean? CSIRO Publishing, Clayton, Australia, 3 edition, 2016.
M. Daher, C. E. G. R. Schaefer, E. I. F. Filho, M. R. Francelino, and E. O. Senra, “Semi-arid soils from a topolithosequence at james ross island, weddell sea region, Antarctica: chemistry, mineralogy, genesis and classification,” Geomorphology, vol. 327, pp. 351–364, 2019.
View at: Publisher Site | Google Scholar
R. J. Schaetzl and S. Anderson, Soils: Genesis and Geomorphology, Cambridge University Press, Cambridge, NY, USA, 2005.
B. Minasny, U. Stockmann, A. E. Hartemink, and A. B. Mcbratney, “Measuring and modelling soil depth functions,” in Digital Soil Morphometrics, A. E. Hartemink and A. B. McBratney, Eds., Springer Nature, London, UK, 2016.
View at: Google Scholar
S. S. Abe and T. Wakatsuki, “The influence of the mound-building termite (Macrotermes bellicosus) on soil clay mineralogy,” Tropics, vol. 22, no. 4, pp. 169–177, 2014.
View at: Publisher Site | Google Scholar
B. B. Mujinya, F. Mees, H. Erens et al., “Clay composition and properties in termite mounds of the Lubumbashi area, D. R. Congo,” Geoderma, vol. 192, pp. 304–315, 2013.
View at: Publisher Site | Google Scholar
P. Jouquet, L. Caner, N. Bottinelli, E. Chaudhary, S. Cheik, and J. Riotte, “Where do South-Indian termite mound soils come from?” Applied Soil Ecology, vol. 117, no. 118, pp. 190–195, 2017.
View at: Publisher Site | Google Scholar
R. W. Borden, I. C. Baillie, and S. H. Hallett, “The East African contribution to the formalisation of the soil catena concept,” Catena, vol. 185, Article ID 104291, 2020.
View at: Publisher Site | Google Scholar
M. E. F. Buss, E. Leizica, R. Peinetti, and E. Noellemeyer, “Relationships between landscape features, soil properties, and vegetation determine ecological sites in a semiarid savanna of central Argentina,” Journal of Arid Environments, vol. 173, Article ID 104038, 2020.
View at: Publisher Site | Google Scholar
J. R. Mcauliffe, M. T. Hoffman, L. D. McFadden et al., “Landscape patterning created by the southern harvester termite, Microhodotermes viator: spatial dispersion of colonies and alteration of soils,” Journal of Arid Environments, vol. 162, 2018.
View at: Publisher Site | Google Scholar
K. Chisanga, E. R. Mbega, and P. A. Ndakidemi, “Prospects of using termite mound soil organic amendment for enhancing soil nutrition in southern Africa,” Plants, vol. 9, p. 649, 2020.
View at: Publisher Site | Google Scholar
P. Merante, C. Dibari, R. Ferrise et al., “Adopting soil organic carbon management practices in soils of varying quality: implications and perspectives in Europe,” Soil and Tillage Research, vol. 165, pp. 95–106, 2017.
View at: Publisher Site | Google Scholar
J. C. Obi, T. O. Ibia, and A. I. Isaiah, “Termite activities and surface characteristics of coastal plain sands,” Journal of Tropical Agriculture, Food Environment and Extension, vol. 18, no. 1, pp. 56–66, 2019.
View at: Publisher Site | Google Scholar
S. Beyene, “Topographic positions and land use impacted soil properties along Humbo Larena-Ofa Sere,” Journal of Soil Science and Environmental Management, vol. 8, no. 8, pp. 135–147, 2017.
View at: Publisher Site | Google Scholar
H. Zhu, W. Hu, R. Bi, D. Peak, and B. Si, “Scale- and location-specific relationships between soil available micronutrients and environmental factors in the Fen River basin on the Chinese Loess Plateau,” Catena, vol. 147, pp. 764–772, 2016.
View at: Publisher Site | Google Scholar
R. S. Macedo, W. G. Teixeira, M. M. Corrêa, G. C. Martins, and P. Vidal-Torrado, “Pedogenetic processes in anthrosols with pretic horizon (amazonian dark earth) in central amazon, Brazil,” PLoS One, vol. 12, no. 5, Article ID e0178038, 2017.
View at: Publisher Site | Google Scholar

Copyright

Copyright © 2023 Abinet Bekele et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

PDF Download Citation

Download other formats

Order printed copies

Views

516

Downloads

472

Citations