Background

Land degradation is a major cause of food insecurity in Ethiopia [1, 2]. Due to increasing human and livestock population pressure, large areas of the country, particularly in the northern highlands, have been exposed to land degradation [2]. Tigray, the northern part of Ethiopia, suffered from extreme land degradation [3]. The rural landscapes of the region were severely suffered from a high degree of soil degradation [4]. Soil and water conservation practices mainly exclosures and stone terraces have been implemented to reverse the land degradation process [5].

Exclosures involve excluding livestock on biophysically degraded communal grazing lands [6] by inhibiting uncontrolled cutting of trees and grass for fuel and fodder [7]. They are effective in regenerating natural vegetation and controlling soil erosion [8, 9].

Stone terraces are another important physical SWC measures to enhance plant growth by conserving moisture and retaining essential nutrients that could have been washed away by erosion [10]. Stone terraces are commonly practiced in steep slo** areas [11] with the participation of farmers. Stone terraces reduced both erosion and sediment transport [12]. Numerous soil properties including soil organic matter and soil aggregation can be improved through application of stone terraces [10].

Establishment of exclosures and construction of terraces enhanced the natural vegetation [13, 14] through improving physical and chemical properties of soils [15, 16]. Aggregate stability is soil physical property considered during restoration of degraded lands [17]. It is a measure of the ability of the soil to resist change due to environmental factors [18]. Soil organic carbon (SOC) and aggregate stability enhance vegetation growth [19]. The stability of soil aggregates influences the water holding capacity of soil and tells the susceptibility of the soil to erosion [20,21,22].

Change in aggregate stability is an indicator of organic matter content, biological activity and nutrient cycling in soils [15] which are essential for the functioning of ecosystems [4]. It is mainly assessed by land management and vegetation recovery [23]. Good understanding of ASD and WSA guides the management of soils against erosive and degradative factors [18]. Land management practices including physical and biological conservation measures improve soil stability to erosion and land degradation [24, 25].

Management practices are determinants in soil aggregation through glomalin [26]. Glomalin is a recalcitrant glycoprotein produced by arbuscular mycorrhizal fungi (AMF) [27, 28]. It protects hyphae during transport of nutrients from the plant to the hyphal tip and from soil to the plant [29]. Its quantification is expressed as easily extractable glomalin and total glomalin [30].

Soil glomalin forms soil aggregates and improves soil structure and stability against erosion [31, 32] as soils with stable aggregates are more resistant to erosion [33]. It is also source of active soil organic carbon [34] and contains 30–40% carbon [35]. The SOC associated with various aggregate size fractions reduces the impact of erosive forces [4] and tells the dynamics of soil organic matter [36].

Measurement of glomalin, organic carbon content and aggregate stability enable to assess the risk of soil structural degradation and function [17, 37]. Previous studies have focused on the effect of land degradation on soil fertility and productivity [9, 34]. However, studies conducted on the effect of SWC measures on soil glomalin, ASD, WSA and AAOC were lacking in the area. Thus, this research was conducted to determine the effect of two decades old CBSWC practices mainly stone terraces and exclosures on soil ASD, glomalin, WSA and organic carbon associated with soil aggregates.

The research questions answered include: Did the construction of stone terraces improved ASD glomalin, WSA and aggregate-associated SOC compared to non-terraced grazing lands? Could exclosure enhanced ASD glomalin, WSA and AAOC compared to terraces and non-conserved grazing lands? and finally could exclosures supported with terraces significantly increase soil ASD glomalin, WSA and AAOC compared to terraces, non-terraced exclosures and non-conserved grazing lands?

Methods

Description of the study area

The study was conducted in Degua Temben district, located 50 km west of Mekelle, regional capital of Tigray region, northern Ethiopia. Geographically, it is located at 13°16′23′′–13°47′44′′ latitude and 39°3′17′′–39°24′48′′ longitude (see Fig. 1). The area has rugged topography. The elevation and morphology are typical for the northern Ethiopian Highlands [39].

Fig. 1
figure 1

Location map of the study area

Full size image

The area receives 290–900-mm rainfall annually with an average value of 615 mm year−1. The rainy season usually occurs between June and September. The highest rainfall is in July and August. The growing season varies between 90 and 120 days. The maximum temperatures occur in May and June.

Soils of the study sites are developed from calcium carbonate-rich parent material [40]. According to World Reference Base [41] soil classification system, Calcaric Cambisols, Vertic Leptosols, Vertic Cambisols and Lithic Leptosols are the dominant soil types. Water erosion is an extremely serious problem; and sheet, rill and gully erosions are observed elsewhere in the study area.

Acacia etbaica, Carissa edulis, Dodonea angustifolia, Stereospermum kunthianum, Rhus vulgaris, Senna singueana and Eucla racemosa are the common woody vegetation species identified in exclosures and in communal grazing lands. Mixed farming system (crop and livestock) is the main livelihood of the study area. Major land uses include forest land, cultivated lands, exclosures and communal grazing lands.

Protection and conservation of the degraded sites involve integrated SWC practices using stone faced terraces, enforcement of grazing restrictions and plantation development [5, 9]. The most commonly practiced SWC measures (i.e., terraces and exclosures with and without terraces) were established since 1997 by the community. Many of the SWC structures constructed are fully owned by the communities. This has contributed toward ensuring their sustainability (BoANR 2014). Before their establishment, the selected SWC measures had a similar history in terms of grazing with the non-conserved communal grazing lands.

Each of the three selected sites (Kerano, Tesemat and Alasa) was categorized into four management units described as terraces, exclosures with terraces, exclosures alone and non-conserved open communal grazing lands (Fig. 2). The management which was classified as exclosure with terrace is restricted from the interference of animals, and both biological planting and physical structures, mainly stone terraces, were implemented. Accumulation of sediments, grasses and litter falls was observed on terraces. More woody plant species were observed compared to the other SWC measures.

Fig. 2
figure 2

Major soil and water conservation measures in one site (a non-conserved open communal grazing land; b terraces; c exclosure + terrace; d exclosure alone)

Full size image

In the case of exclosures without terraces, there was no interference of livestock and human practices. Besides, no other management practices such as physical structures were observed. Trees regenerate naturally and hence better vegetation cover than terraces. Erosion types such as sheet, rill and gully formation are relatively less common compared to non-conserved grazing lands.

The stone terraces in grazing lands were selected as third management unit because they are relatively more stable physical SWC measures. It was observed that terraces had better sediment deposits and vegetation cover. Besides, sheet, rill and gully erosion types were less common compared to the open non-conserved communal grazing lands.

The open, non-conserved communal grazing land was characterized by the low vegetation cover and higher proportions of bare soil with high stone cover. Sheet, rill and gully erosion were very common. It is assumed that the terraces, exclosures and non-conserved grazing lands had comparable initial conditions at the time of terrace construction and exclosure establishment. Changes in soil ASD, WSA and AAOC were assumed to be as a consequence of terraces and exclosures establishment.

The area coverage of terraces ranged from 13.87 to 24.42 ha and that of exclosures + terraces and exclosures alone ranged 12.74–51.80 and 14.02–34.70 ha, respectively, while area of the adjacent non-conserved communal grazing lands ranged 11–34.96 ha.

Sampling technique and sample size

The study was conducted in three nearby sites (Kerano, Tesemat and Alasa with in the district) and having all the SWC measures. In each SWC measures, three transects separated at a minimum distance of 75 m were established. Transects were parallel to each other and to the topography of the landscape. In each transect, three landscape positions (i.e., upper, middle and foot slope) were established. The upper slope (US) position is the uppermost portion of each study site, and it can receive little or no overland flow but may contribute runoff to down slope areas. The middle slope (MS) position receives overland flow from the upper slope and contributes runoff to the foot slope (FS). The FS represents the lowest part of each study site and receives overland flow from both mid- and upper slopes [42].

Soil samples were collected from 0 to 30 cm at four corners and center of a 10 m × 10 m size plot using “X” sampling design from terraces, exclosure + terrace, exclosure without terraces and from non-conserved communal grazing lands. A total of 108 soil samples (i.e., four conservation measures*three slope positions*three samples*three replications) were collected for glomalin, ASD, WSA and AAOC determination.

Extraction and determination of glomalin-related soil proteins

The method described by Wright and Upadhyaya [31] was used to determine the easily extractable and total glomalin-related soil proteins. A one-gram sample of air-dried soil was placed in 8 mL 20 mM citrate, pH 7.0, and autoclaved (121 °C) for 30 min to remove the easily extractable glomalin (EEG). After centrifugation (10,000×g) and removal of the supernatant, 8 mL 50 mM citrate, pH 8.0, was added to the remaining soil and heated at 121 °C for 60 min to extract total glomalin (TG). Extractions continued with 50 mM citrate until the supernatant becomes straw in color, indicating that glomalin, a red-brown color, had been removed. One mL of EEG was removed, and then, the remaining supernatant containing EEG was combined with all of the supernatants from the 50 mM citrate extractions. Bradford dye-binding assay was used to determine protein with bovine serum albumin as a standard [43].

Aggregate stability determination

The method described by Kemper and Rosenau [44] was used to determine water stability of air-dried aggregates. Hundred grams of air-dried bulk soil that passed through 8-mm sieve was sieved using 5-, 2-, 1-, 0.5-, 0.25- and 0.053-mm sieves. Each fraction was pre-wetted overnight by capillary action and then transferred on the top of a nest of the same sieves immersed in water. The nest of sieves was then vertically tumbled in a column of water for 5 min, at a rate of 50 complete repetitions per minute. The mass of oven-dried particles (105 °C for 24 h) that resisted breakdown was assessed for each sieve. The respective dry masses were used to compute the mean weight diameter (MWD) and water stable aggregate (WSA) as follows:

$${\text{WSA}} = \left( {(M\left( {a + s} \right) - {\text{MS}}} \right)/\left( {{\text{Mt}} - {\text{Ms}}} \right)) \times 100$$
(1)

where M(a + s) is the mass of resistant aggregates plus sand (g), Ms is the mass of resistant aggregates and Mt is total mass of soil

$${\text{MWD}} = \sum {\text{X iW i }}$$
(2)

where MWD is the mean weight diameter of water stable aggregates, ** is the mean diameter of each sieve fraction (mm) and Wi is the proportion of the total sample mass in the corresponding size fraction

$${\text{Stability index }}\left( {\text{SI}} \right) = \frac{1}{{\left( {\text{MWD}} \right)d - \left( {\text{MWD}} \right)w}}$$
(3)
$${\text{SQ}} = {\text{SI}} \times {\text{percentage of aggregates}} > 2\;{\text{mm}}$$
(4)

where SQ = stability quotient.

Determination of aggregate-associated organic carbon

Soil organic carbon content was determined by Walkeley and Black [45] after sieving with 0.25-mm sieve for micro-aggregates and 2-mm sieve for macroaggregates.

Data analysis

Analysis of variance (ANOVA) was used to see for any significant differences of the parameters along the different community soil and water conservation measures using SAS 9.2. Mean comparison was carried out using Duncan’s multiple range test (DMRT), and finally, correlation analysis was carried out to see the relationship between ASD, WSA, glomalin-related proteins and soil organic carbon associated with macro- and micro-aggregate fractions.

Results and discussion

Effect of CBSWC measures on soil glomalin

The content of glomalin under the different CBSWC measures showed that EEG was significantly (p < 0.05) higher in exclosures compared to non-conserved grazing lands. Terraces also had relatively higher EEG compared to non-conserved grazing lands. Easily extractable glomalin in exclosures was 11–27.93% higher than on terraces and 44.65–55.17% higher than that of non-conserved grazing lands. Besides, EEG on terraces was 37.79% higher than non-conserved grazing lands. The order of EEG was in the order of exclosures + terraces > exclosures alone > stone terraces > non-conserved grazing lands (Table 1). The presence of higher glomalin in exclosures could be due to the presence of high AMF root colonization as glomalin is produced by mycorrhizal  fungi. It was reported that glomalin stocks are greater where AMF is more abundant [46].

Table 1 Soil glomalin under the different SWC measures in mg g−1 soil
Full size table

Total glomalin was also significantly (p < 0.05) higher in exclosures compared to non-conserved grazing lands. Total glomalin in exclosures was 37–45% higher than in non-conserved grazing lands and 24.55–35.51% higher than that of terraces. Livestock grazing could be the cause for low content EEG and TG in non-conserved grazing lands. It was reported that grazing and trampling had negative effect on the amount of glomalin through decreasing vegetation cover [47, 48]. Furthermore, as glomalin is produced by AM fungi, trampling may have destroyed the aggregate of the soil and break the hyphae. It was also reported that other factors such as landscape characteristics can affect the amount of glomalin [49].

Effect of soil and water conservation practices on aggregate size distribution

The result of aggregate size distribution indicated large macroaggregates (> 2 mm) were higher than other aggregate sizes in all SWC measures. This is similar to the result of **ao et al. [50] who found large macroaggregates (> 2 mm) represented the greatest fraction for all the land uses considered. The effect of different SWC measures on dry aggregate size distribution (Table 2) indicated that exclosures + terraces had significantly higher percentage (32.41%) in the > 2 mm fraction. This could be due to the presence of high organic carbon and low disturbances in exclosures. High organic matter from litter falls and decayed tree roots increased coarser aggregates [51]. Besides, it was reported that strong plant root systems are beneficial for soil aggregation [23]. This also agreed with the result of [36] who found higher macroaggregates in vegetated area than bare land.

Table 2 Percentage of dry aggregate size distribution under the different SWC measures (mean ± SEM)
Full size table

Non-conserved communal grazing lands had significantly lower (21.91%) in this size class. Destruction of macroaggregates by livestock trampling and soil erosion may be the cause for the decrease in large macroaggregates in non-conserved communal grazing lands. Furthermore, clearing of the natural vegetation could be the cause for the dispersion, detachment and decrease in macroaggregates as vegetation cover protects the soil from detachment and aggregate breakdown. The order of percentage of aggregates (> 2 mm fraction) was exclosures + terraces > exclosures alone > terraces > non-conserved communal grazing lands.

Though not significant, terraces had relatively lower percentage (18.15%) of aggregates in 1–2 mm fraction than the other SWC measures. The decrease could be due to the deposition fine sediments on terraces. However, exclosures + terraces had significantly higher (22.49%) which could be due to the presence of high organic carbon. Soil aggregation has positive and strong relation with soil organic carbon [53]. The decreasing order of percentage of aggregates (1–2 mm fraction) was: exclosures + terraces > exclosures alone > non-conserved communal grazing lands > terraces.

Exclosures with terraces had significantly (p < 0.05) higher percentage (12.61%) of aggregates in the 0.25–0.5 mm fraction followed by terraces. This indicated terraces in exclosures and in grazing lands decreased detachment and dispersion of soil particles by erosion. The presence of lower (10.41%) of aggregates in this size class in the non-conserved communal grazing lands could be due to soil dispersion by erosion. The decreasing order of percentage of aggregates in the 0.25–0.5 mm fraction was: exclosures + terraces > terraces > exclosures alone > non-conserved communal grazing lands. This revealed high disturbance in non-conserved grazing land has deteriorated soil structure. It was reported that soil aggregates destruction depends on the degree anthropogenic disturbances [52].

Next to the 2–5 mm size fraction, the highest percentage (26.66%) was found in the < 0.25 mm fraction (micro-aggregates). Non-conserved grazing lands had the highest percent of aggregate in the < 0.25 mm size. This could be due to dispersion of aggregates by disturbance. This could indicate soil particles are more likely to be detached in non-conserved grazing lands. In connection to this, Singh et al. [36] found higher micro-aggregates in bare land. The decreasing order of percentage of micro-aggregates (< 0.25 fraction) was: non-conserved communal grazing lands > stone terraces > exclosures alone > exclosure + terraces.

Effect of CBSWC measures on water stable aggregates

Wet aggregate stability values varied among the CBSWC measures (Table 3). It was significantly higher (p < 0.05) in exclosures. The presence of higher soil macroinvertebrates density and mycorrhizal association in exclosures could have improved soil aggregation in exclosures [25]. Especially, termites and earthworms are known to enrich the soil with organic materials and improve soil structural stability [27]. Mycorrhizal fungi entangle particles within the hyphae network and cement particles together [35]. Besides, the presence of high vegetation cover in exclosures increased supply of organic matter inputs and decreased soil erosion. In line with this, Fokom et al. [32] observed a decrease in WSA as forest is converted to other land uses. This supports the idea that greater stability is associated with organic matter supply [52] because organic matter improves establishment of soil structure through binding and limits soil erosion [15, 20]. Vegetation has also mulching effect to improve soil aggregation [53, 54].

Table 3 Effect of SWC measures on aggregate stability (mean ± SEM)
Full size table

Communal grazing lands had significantly (p < 0.05) lower percent of water stable aggregates. This could be due to physical disturbance and low soil organic carbon. Grazing decreased aggregate stability [20] through dispersion of soil aggregates [15].

Conversion of communal grazing lands to exclosures resulted in 20–21% increase in percent of WSA, and conserving the open communal grazing lands with terrace resulted in an increase in WSA by 12%. The WSA decreased in the order of exclosures + terraces > exclosures alone > terraces > non-conserved communal grazing lands. The result of stability quotient also indicated soils in exclosures are 1.23–1.6 times more stable than non-conserved grazing lands. Terraces are also 1.27 times more stable than non-conserved grazing lands (Table 3). This indicated exclosure areas showed relative recovery from structural degradation. It could also indicate less soil erodibility in exclosures. Vegetation cover in exclosures protected the soil from structural disturbance [54]. However, grazing lands due to their bare surface, they receive few inputs of organic matter and are susceptible to degradation. Grazing breaks the soil apart, exposing the organic matter to microbial decomposition and facilitates soil loss by erosion. It was found that due to trampling effect, bulk density increased in free grazing lands, while aggregate stability decreased [51].

Mean weight diameter of aggregates is one measure of aggregate stability [18]. Exclosures had significantly higher (p < 0.05) in MWD than the non-conserved communal grazing lands. The MWD of the different SWC measures follows decreasing order of exclosures alone > exclosures + terraces > terraces > non-conserved communal grazing lands. This revealed higher organic matter in exclosures stabilized the soil through aggregation. High values of MWD could indicate lower erodability of soils in exclosures. It would also imply stable aggregates are critical to erosion resistance. It was found that establishment of exclosures on degraded lands restores aggregate stability [55]. It was also reported that vegetation cover improved soil structure [15]. Mean weight diameter was found to be highly responsive to cover [56]. This indicated organic matter increased both WSA % and MWD through binding [57].

Bareness and disturbance by grazing are the causes for dispersion of soil aggregates [58], and this might be the cause for the decrease in MWD in non-conserved grazing lands. Similar study reported that the main mechanism of aggregate breakdown is by dispersion through disturbance [51]. Mean weight diameter was responsive to livestock grazing because it was reported that MWDs were lower in grazed grassland area [20].

Effect of SWC measures on aggregate-associated soil organic carbon

Soil and water conservation measures significantly (p < 0.05) affected the soil organic carbon associated with macroaggregates (> 0.25 mm size). However, no significant variation was observed on those associated with micro-aggregates (< 0.25 mm) (Table 4).

Table 4 Aggregate-associated organic carbon under the different SWC measures (mean ± SEM)
Full size table

Exclosures + terraces had significantly (p < 0.05) higher (3.1%) organic carbon associated with macroaggregates, while non-conserved communal grazing lands had lower (2.2%) macroaggregate-associated carbon. Shrestha et al. [59] found higher amounts of associated SOC concentration under undisturbed sites. **ao et al. [50] reported the highest SOC in large aggregates under exclosures. In our study, the conversion of communal grazing lands to exclosures resulted (17–27%) increase in SOC in macroaggregates followed by terrace (9%) as compared to non-conserved grazing lands.

The highest (2.8%) organic carbon associated with micro-aggregate was found in exclosures + terraces, while the lowest (1.8%) was found in non-conserved communal grazing lands. Conversion of communal grazing lands to exclosures resulted in 19–34% increase in SOC in macroaggregates, while terrace construction resulted in 17% increase in macroaggregate carbon as compared to non-conserved grazing lands. The low aggregate carbon in non-conserved communal grazing lands could be due to low biomass input caused by livestock grazing and human disturbances.

Soil organic carbon associated with macroaggregates was higher than SOC associated with micro-aggregates. This indicated macroaggregate structures are important in physical protection of soil organic carbon. In line with this, Gelaw et al. [4] in Mandae watershed, northern Ethiopia, found higher SOC in macro- than micro-aggregates. Many other authors also reported that organic carbon in coarse aggregates was higher than in fine aggregates [17, 50, 60]. This was due to high rate of decomposition in micro-aggregates [61]. This suggested that micro-aggregate stability could be a better indicator of potential soil erosion hazards [36]. A large decrease in SOC from macro- to micro-aggregate (16%) was observed in exclosure with terraces followed by non-conserved communal grazing lands (16%) and exclosures with terraces (12%), while the lowest (5%) was observed in terraces.

The SOC associated with both macro- (> 0.25 mm) and micro-aggregates (< 0.25 mm) follows the order exclosures + terraces > exclosures alone > terraces > non-conserved communal grazing lands. This showed the deposition and turnover of litter fall, stump and roots of matured trees maintained SOC in exclosures. This implied that higher SOC content would further contribute to soil aggregate stability in the restoration of grazing lands [62].

Relationship between soil glomalin, organic carbon fractions, MWD and aggregate stability

This study found positive and significant (p < 0.05) relationship between EEG and water stable aggregates. Easily extractable glomalin and total glomalin explained about 43 and 40% of the variation in percent of water stable aggregates, respectively (Fig. 3). The relationship between TG and % WSA was also positive and significant. In line with this, a study by Sirinikorn et al. [49] and Hontoria et al. [63] found positive relationship between soil aggregate stability and glomalin content. Similarly, Wright et al. [64] reported that glomalin-related soil protein increased as aggregate size increased. This indicated both EEG and TG are important for soil aggregation and enhance stability. This could be due to the cementing and recalcitrant properties of glomalin. It was reported that glycoprotein, produced by AMF hyphae, has a cementing capacity to maintain soil particles together [29]. It also has relatively slow turnover in soil, contributing to lasting effects on aggregation [27].

Fig. 3
figure 3

Relationship of %WSA with EEG and TG

Full size image

The relationship between EEG and % SOC and TG with SOC was positive and significant. Easily extractable glomalin and TG explained 23 and 21% of the variation in % SOC, respectively (Fig. 4). These positive and significant relationships indicated that EEG and TG contribute to SOC storage. Zhang et al. [65] reported glomalin-related soil proteins are used for preserving and accumulating SOC. Glomalin contains approximately 30–40% of carbon and forms small soil clumps. This granulated material binds carbon in the soil [59].

Fig. 4
figure 4

Relationship of % SOC with EEG and TG

Full size image

The positive relationship between aggregate-associated soil organic carbon and water stable aggregates (Figs. 5, 6) indicated SOC is necessary in soil aggregation. This agreed with the result of Dorji et al. [53] in a montane ecosystem of Bhutan and Gelaw et al. [4] in Mandae watershed of the northern highlands of Ethiopia. This could be due to the cementing and recalcitrant properties of glomalin which significantly enhance the stability of soil aggregates through SOC sequestration [27, 64, 66] and slowing its turnover [40]. It was reported that glomalin is glycoprotein nature and hydrophobic characteristics; therefore, it is a very stable biomolecule, with a half-life in soil between 6 and 42 years, and prevents nutrients losses [67].

Fig. 5
figure 5

Relationship of %WSA with micro- and macroaggregate carbon fractions

Full size image
Fig. 6
figure 6

Relationship of MWD with micro- and macroaggregate carbon fractions

Full size image

Conclusions

Identifying sustainable soil and water conservation practices is necessary to solve land degradation. The application of exclosures and terraces increased soil aggregate stability. Soils with high glomalin content had high organic carbon and water stable aggregates. Larger proportions of macroaggregates were found in exclosures followed by terraces, while large proportions of micro-aggregate fractions were found in non-conserved communal grazing lands. Aggregate stability increased with organic matter content. Macroaggregates contain higher SOC than micro-aggregates. Significantly higher water stable aggregates were found in exclosures compared to terraces and non-conserved grazing lands. Implementation of exclosures and terraces should be expanded to enhance the aggregate stability of the soil and organic carbon associated with aggregates and finally to rehabilitate soil structure degradation.