1. Introduction
Understanding the interplay between land use systems and soil properties is key to find avenues to sustainable agricultural production [1]. Vertisols generally show a large spatial range in soil properties, even over relatively short distances. As a result, they remain the most difficult land resource systems in the world to manage successfully [2,3]. The wide variability in vertisol is due to its particular clay mineralogy, which can cause shrinking and major water-transporting bypass flow cracks in dry conditions and churning and swelling properties, leading to local waterlogging in wet conditions. Often, access to the land for mechanized tillage is limited to particular time slots. Land use and management practices then also add to the spatial variation.
Ethiopia, with a land area of 1.13 million km2, is characterized by considerable diversity in terms of its biophysical environment, geology, topography, agro-ecology and soils [4]. This diversity, in turn, has resulted in the recognition of 18 soil classification orders [4,5]. Figure 1 shows that vertisols are among the most extensive soils, accounting for about 11% or 12.6 million ha of the Ethiopian landmass. A more recent survey in the highlands found vertisols to cover about 27% of the agricultural area [4]. When mapping soils and land use systems in the Ethiopian highlands, vertisols turned out to be distributed as 36%, 31%, 11% and 22% in the land use systems FS1, FS2, FS3 and FS4, respectively. Figure 1 shows where the two overlap.
Vertisols are particularly extensive in the volcanic plateaus and the colluvial slopes of the north-central highlands characterized by a wheat–teff –maize–livestock (teff [Eragrostis teff (Zucc)] is a small cereal extensively cultivated in Ethiopia for its grain that is used to make the favorite national food—injera) farming system; the sandstone colluvium of the southern Tigray characterized by a sorghum–teff–livestock system; the granitic and rhyolitic colluvium of the southern highlands characterized by an enset–coffee–cereal–livestock system; and the limestone colluvium of the Hararghe plateau that corresponds with the khat–sorghum–livestock system (Khata edulis—A bushy tree plant that is cultivated for its succulent leaves that are consumed as stimulant throughout the region and exported to Djibouti and Somaliland), as shown in Figure 1 [4].
Unlike specialized cropping systems, these mixed systems are complex agricultural arrangements that integrate crop and livestock production at the farm or landscape level with differing degrees of interaction among system components. The common vision is that soil properties, translated into soil quality, drive crop production, but land use systems at some point also determine spatial soil quality differences, taking the shape of ‘soil phenoforms’ [6,7].
A voluminous amount of literature has been produced about the physical, chemical and mineralogical properties of vertisols and related management challenges in the Ethiopian highlands [4,8,9,10]. However, the variability of vertisol properties across spatial scales and as a function of land use practices and farming systems trajectories is difficult to grasp [11,12]. This is aggravated by gully erosion, and the depletion of organic matter and nutrients (mainly N and S), accompanied by the deterioration of soil structure and increased bulk density, including salinity build up in the irrigated lowlands [4].
Ethiopian farmers have a long-standing tradition of vertisol management. In addition to the soil burning (guie), early planting of crops with short growing periods and late planting to take advantage of the residual moisture after the main rainy season stops, are common to farmers in the highland areas. An example is wheat followed by chick pea in a relay cropping system. Drainage furrows made with the maresha, the traditional plough, are also common practices to drain excess water from farmlands. The broad bed maker (BBM) is the contribution of the scientific community researching on the management of vertisols in the county. It was first introduced by The International Crops Research Institute for the Semi-Arid Tropics (ICRISAT) in the 1970s [13]. The purpose was to produce broad beds and furrows with oxen-drawn implements, reducing the labor burden on farmers. Through continued research, BBM has become more efficient and affordable for farmers. Later on, the ‘BBM package’ was developed, that includes moisture management, appropriate planting time, selection of improved crop variety and fertilizer recommendations to increase crop yield in vertisol-dominated areas of the country [14].
Such developments are good, as the vertisol area is key to the country’s future food security. Out of 17 million rural households in Ethiopia, not less than 14.5 million have access to fertilizers. At the same time, some 10.5 million households have average land holdings of below 0.6 ha, at an average household of five family members [15]. This clearly shows the need to apply fertilizers in a way that brings about yield increases that are beneficial to farmers to achieve food self-sufficiency. On vertisols, fertilizer use is by far low, and presented as one of the major causes of low wheat (2.3 t/ha) and teff (1.3 t/ha) yields in the Ethiopian highlands [16]. Soil fertility management has been based on the application of combinations of DAP (18%- N & 46%- P2O5) and urea (46% N). More recently, however, DAP has been replaced by a compound fertilizer (NPS) with additions of micronutrients. Currently, the most popular blend being promoted through the agricultural extension system is the Zn-B blend (NPS + ZnB) with ratios of 17 N-34 P2O5 + 7 S + 2.2 Zn + 0.6 B. As this change of focus is rather recent, its added benefits still remain to be proven [17].
The present study was part of the larger CASCAPE project (Capacity Development for Scaling Agricultural best practices in Ethiopia), which was meant to support Ethiopia’s Agricultural Growth Program. The project conducted extensive soil survey to characterize and map major soil landscapes in the Ethiopian highlands during the 2014–2015 period [4,18]. This paper is set out (i) to look for correlations between Ethiopian highland land use systems and vertisol characteristics; and (ii) to examine crop response to fertilizers applications on vertisols (mainly wheat and teff) and to derive optimum rates of fertilizers application for wheat and teff production on vertisols in the Ethiopian highland land use systems.
2. Materials and Methods
2.1. Soil Analytical Data
The analytical data on Vertisols were obtained from CASCAPE’s soil profile that was generated as part of a major soil survey in the Ethiopian highlands [4,18], which led to several district-level soil maps [19]. In order to explore the vertisol spatial variability, analytical data from the topsoil (0–20 cm) of 45 vertisol profiles were collected across the different land use system zones, out of a total of 204 georeferenced mapping units. Selected physical (particle size distribution, bulk density) and chemical properties (soil pH, organic carbon (OC), total nitrogen (TN), available phosphorus (AP), cation exchange capacity (CEC), exchangeable bases (Ca, Mg, Na, and K) and micronutrients (Fe, Mn, Zn, Cu, and B)) were considered.
Soil analysis was performed at the soil fertility laboratory of Waterworks Construction and Design Supervision Enterprise in Addis Ababa, applying standard laboratory procedures as outlined by the International Soil Reference and Information Centre [20]. Soil pH-H2O was measured using 1:2.5 soil to solution suspension using a pH meter, and the Walkley and Black method was used to determine OC content. The TN content was determined using the Macro–Kjeldahl method [21], while AP (Olsen) was measured using sodium bicarbonate extraction solution [20]. The CEC (cmol (+)/kg) was determined by the ammonium acetate method; exchangeable bases and S were determined using Mehlich-3 [22]. The contents of available micronutrients were extracted using the DTPA extraction method [23].
2.2. Fertilizer Trial Design
Teff trials were conducted during the 2017 and 2018 cropping seasons, while wheat trials were conducted in 2018 on vertisols. Teff trials were conducted in Burie and Becho districts representing FS2 and the Enamore district representing FS3. Table 1 presents some topographic features (elevation, and slope) and climate data (mean annual rainfall, mean min and mean max temperatures) of the study sites.
Wheat trials were conducted in Endamohoni and Girar Jarso districts representing FS1 and FS2 (Figure 1). The trials involved seven treatments: five levels of NPS + ZnB blend fertilizer (50, 100, 150, 200 and 300 kg/ha), 100 kg/ha NPS, and 150 kg DAP/ha, which is the formerly recommended blanket application rate for cereal on vertisols. This allows production function analysis for the blend fertilizer, as well as comparing the full blend and NPS at 100 kg/ha, and DAP and NPS at the 150 kg/ha level.
Treatments were laid out in a randomized complete block designs (RCBD) using three farm fields as replications on plot sizes of 5 m × 5 m. The seedbed was prepared by ploughing the land four times using oxen-drawn implements, and planting was carried out in mid-July in both years. Improved varieties of teff (Kuncho in Bure and Becho, and Boset in Enamor) and wheat variety (Hidassie) were used. Wheat was planted in rows at seeding rate of 125 kg/ha, while teff was broadcast at a seeding rate of 10 kg/ha. All plots received 100 kg/ha urea that was applied in three splits—one third each at planting, two weeks after emergence and at booting stages. At crop maturity, a 2 m × 2 m plot was harvested to the ground level to determine the biomass and grain yield per plot. Rainfall data (Table 1) were obtained from nearby meteorological stations in the trial sites.
2.3. Data Analysis
The soil profile data were subjected to statistical analysis to produce descriptive statistics (mean and variance) and analysis of variance using R-statistical software [24]. Given the uneven distribution of vertisols across different land use systems, ANOVA for unbalanced design was employed. As a result, type III sum of squares was requested, using the car package of R [25] while performing the ANOVA. The crop yield data from fertilizer trials were subjected to ANOVA as well. When the ANOVA resulted in a significant difference, mean separation was achieved by means of Tukey’s honestly significant difference (HSD) at the 5% level. This explored how variation among individual soil properties and crop yields were related to each other. Soil analytical data for the different land use systems were used to perform a correlation analysis to explain the biomass and grain yields to the test crops.
3. Results and Discussion
3.1. Variability in Biophysical Properties and Soil Management across Land Use Systems in the Ethiopian Highlands
The dominant in situ parent material in the north-central highlands is olivine basalt and recent pyroclastic deposits and pumice (mainly along the western horst arm of the Rift Valley). Unwelded tuff and lacustrine sediments with basaltic rocks at the base predominate in the southern highlands. In the depressions and foot slopes where vertisols are dominant, colluvial and alluvial deposition of transported materials enrich the soil with clay content [8]. The altitude is between 1500 and 3000 m, while the temperature ranges between 10 and 26 °C, creating tepid moist to sub-moist mid-highlands agro-climatic conditions in the middle altitudes. Rainfall is highly variable with the western half of the country (i.e., the north-central and south-western highlands) receiving high annual rainfalls (800–1400 mm), while the eastern half has dry and hot climates (Table 1). The seasonal distribution of rainfall is an important feature in Ethiopia, marked with alternating wet and dry seasons (Figure 2).
The soil moisture regime in the southern Tigray region was identified as ustic, while the north-central and southern highlands were characterized as udic and perudic, respectively [26]. Similarly, the soil temperature regimes were classified as thermic for the north-central and southern highlands, and hyperthermic for the southern Tigray plateau.
In the sorghum–teff–livestock system (FS1), low rainfall and moisture stress is a key constraint to produce sorghum (Sorghum bicolor), teff and some vegetables. Hillside terraces and stone bunds are used for soil and water conservation. In the FS2 areas, Vertisols are used for cultivation of wheat (Triticum aestivum) and teff. Chickpea (Cicer arietinum), lentil (Lens culinaris) and grass pea (Lathyrus sativus) are grown in the dry season as catch crops using the residual moisture. In order to drain the excess water and tackle poor soil workability during rainy season, farmers practice soil burning (locally known as gueing). The practice alters soil physical properties (i.e., soil color changes into redder hues, fused clay turns into sand-sized particles) and chemical properties, including the destruction of organic matter and loss of N and S contents, but it results in a considerable increase in the contents of AP, exchangeable K and soil pH [4,27].
In addition, free-range grazing is practiced in the highland cereal–livestock systems (FS1 and FS2) in the cropping fields. Livestock are driven into crop fields for aftermath grazing on crop stubble, weeds, and other vegetation, which results in compaction through trampling the soil. This leaves the soils eventually bare, and often leads to massive erosion at the onset of the rainy season. Further, animal manure is turned into “dung cakes” and used as household fuel, denying the soil an important source of organic matter and nutrients (Figure 3). The opposite holds true for FS3 and FS4, where livestock is managed in cut-and-carry feeding systems due to the perennial nature of cultivated crops and enrichment of the soil with farmyard manure.
3.2. Variability in Vertisol Properties across Different Farming Systems
Descriptive statistics and mean variation of soil properties across the four farming systems are presented in Table 2. A high coefficient of variation (CV) was observed for some soil parameters (e.g., OC, TN, and AP) that are largely influenced by the variability in land use practices presented above. In particular, variations in crop (perennial vs. cereal), livestock management (free range vs. cut-and-carry) and soil management practices (e.g., crop residue and dung burning, soil burning, and fertilizer rates) explain the variations in soil properties studied. On the contrary, variations of certain soil properties, such as high clay content, neutral to slightly alkaline soil pH, high CEC and exchange complex saturated with Ca and Mg, are reflections of the inherent properties of vertisols [3].
The analysis of variance shows highly significant differences (p < 0.001) among farming systems in particle size distribution (mainly sand fraction), soil pH, OC, TN, AP and micronutrients, except for Cu (Table 2). Although vertisols are known for their high clay content, there was significantly higher sand fraction under FS2 (33%), compared to that under FS3 (15%). This is associated with the practice of soil burning that is common in FS2. Guieing is reportedly known to turn fused clay into a sand-like structure. Similar findings are reported elsewhere and for other soil types [27,28,29]. Bulk density (BD) is significantly higher in soils under FS1 (1.44 g/m3) and FS2 (1.26 g/m3) where land is intensively cultivated, and free grazing is common. This is in sharp contrast with the BD value (1.05 g/m3) observed under FS3, where the hoeing and cut-and-carry feeding of animals is practiced.
Soil pH ranges from strongly acidic (5.58) under FS2 to neutral (7.35) under FS4. Whereas the strongly acidic reaction may be explained by continued application of N and P fertilizers, the high pH under FS4 is the reflection of high Ca saturation and the frequent occurrence of calcareous alluvial deposits that characterize this land use system [30]. The mean values of OC (5.58%) and TN (0.51%) are rated as high for vertisols, which are usually characterized as low in these properties due to intense mixing of organic matter with clay complexes under anaerobic conditions during wet seasons [3]. The significantly (p < 0.001) higher mean OC (13%) and TN (1.42%) under FS4 is attributed to the accumulation of khat leave residues that are dumped on crop fields after the succulent parts are consumed as a stimulant. In the same manner, the high OC and TN levels under FS3 are due to organic matter enrichment through the application of farmyard manure (FYM) and household refuse, which is the characteristic feature of the enset–coffee system [18]. The intensely used ‘annual crop’ land use systems FS1 and FS2 have a much lower OC of 1.5 to 1.9%, which is still much higher than the African average of 0.8% [12,31,32]. The mean value for AP (11 mg/kg) is in the medium range, showing highly significant (p < 0.01) variation between farming systems. The highest mean AP value (18 mg/kg) was observed in FS1 and the lowest (7 mg/kg) in FS2.
There was highly significant (p < 0.01) variation in the levels of CEC and exchangeable bases, except for Na that was present only in trace amounts in the soils under almost all farming systems. The highest level of CEC (53 cmol(+)/kg) was observed in FS3, followed by FS4 (51 cmol(+)/kg), which coincides well with the high clay and OC contents in these farming systems. The highest Ca (33 cmol(+)/kg) and Mg (13 cmol(+)/kg) contents were observed in FS4 and FS2, and the lowest in FS3 (Table 2), which can be explained by variations in the parent materials between the farming systems. The soils in FS4 are derived from Ca-rich calcareous materials (chiefly limestone), and those in FS2 are derived from alkali basalt rocks which are rich in Mg [30]. In contrast, silica-rich parent materials (e.g., rhyolites and trachytes) are dominant under FS3.
Furthermore, the micronutrient levels are significantly (p < 0.001) varied across farming systems with the exception of Cu, reflecting differences in land use and parent materials. The levels of Fe and Mn are particularly high under FS2, consistent with the strongly acidic soil reaction under this farming system (Table 2). At higher pH, the precipitation of micronutrients is also possible (i.e., cations are strongly bound by hydroxides or oxides), thereby reducing their availability in the soil solution [33].
3.3. Teff and Wheat Grain and Biomass Yield and Their Correlation with Soil Properties
Table 3 presents teff and wheat yields, averaged over years (for teff) and over all fertilizer treatments. The data showed highly significant (p < 0.001) variations in teff and wheat grain and biomass yield across land use systems. The highest teff grain yield (2.15 t/ha) was obtained at Becho in FS2, where the biomass yield (9.17 t/ha) was the lowest. Conversely, the lowest grain yield (1.34 t/ha) but with the highest biomass yield (12.47 t/ha) was obtained at Enamore in FS3, showing an inverse relation between biomass and grain yields of teff. The inverse relation between teff grain and biomass yield is also clearly indicated in Figure 4 for the two cropping seasons. This is meaningful to farmers, as a poor yield may still have the benefit of high stover production, which is needed to feed the oxen during the dry season.
In the case of wheat, the highest plant height (133 cm), biomass (14.03 t/ha) and grain (5.78 t/ha) yields were obtained at Endamohoni in FS1. This is more than twice the national average (2.25 t/ha) and three times higher than the yield obtained at Girar Jarso in FS2 (1.95 t/ha). The large differences, particularly for wheat, indicate that fertilizer application alone is not a sufficient condition to attain higher yields as the soil type and its conditions (e.g., OM, nutrient content, and water logging), along with climate (rainfall) can be sometimes more significant factors. For example, as shown in Figure 2, the Girar Jarso site in 2018 received persistently high rainfall throughout the June–August period, unlike the Endamohoni district. As result there was a temporary water logging problem in the Girar Jarso district, which affected plant growth and biomass yield. In the same line, almost all soil fertility parameters evaluated were significantly higher at the Endamohoni site that does not experience waterlogging problems.
The results of the Pearson correlation analysis between individual soil properties and crop yields are shown in Table 4. Teff biomass was significantly and positively correlated to OC (r = 0.37), S (r = 0.32), Fe (r = 0.43), Mn (r = 0.47) and Cu (r = 0.29), but these soil properties were negatively correlated to teff grain yield except for Cu. The teff grain yield was negatively correlated to soil pH (r = 0.32), exchangeable K (r = 0.44), and B (r = 0.23). On the other hand, there was a negative correlation (r = −0.11) between biomass and grain yields of teff, which can be attributed to the problem of lodging. When a teff plant grows taller and produces large biomass, the weak stem cannot support it, and lodging occurs.
Unlike teff, there was a positive and highly significant (p < 0.01) correlation between wheat biomass and grain yields that followed a similar pattern to soil individual properties. Wheat grain yield was significantly and positively correlated with soil pH (r = 0.56), OC (r = 0.86), TN (r = 0.68), AP (r = 0.53), Cu (r = 0.78) and B (r = 0.60), while exchangeable K (r = −0.84) and S (r = −0.43) were negatively and significantly correlated. The correlation pattern is almost the same for wheat biomass yield (Table 4).
For the three nutrients in the blend fertilizer, only S for teff biomass and B for both wheat grain and biomass correlated significantly. This correlation analysis would, therefore, not support their inclusion in the fertilizer.
3.4. The Type and Level of Fertilizer to Apply on Vertisols
Averaged over years and locations, the analysis of variance for grain and biomass yield across different levels of blend fertilizer treatments showed highly significant variations (p < 0.01) for treatment effects (Table 5). The highest level of blend fertilizer (300 kg/ha NPS + ZnB) produced the highest biomass and grain yield for wheat, but it depressed the teff grain yield, which is negatively correlated with higher biomass yield (Table 4). For teff, the added yield advantage of going from T1 to T5 was 3.43 t/ha biomass (38% increase) and 0.48 t/ha for the grain (33% increase). On the other hand, comparison between T1 and T2 revealed a yield difference of 2.2 t/ha (25% increase) for teff biomass and 0.17 t/ha for the grain (25% increase). Beyond T2, the grain yield tends to level off, suggesting that 100 kg/ha NPS + ZnB is a recommendable rate of application for teff grain production on vertisols in the Ethiopian highlands.
In contrast, the highest wheat yield advantage (4.72 t/ha), which is a 65% increase in biomass over the control plot and 2.38 t/ha and 90% increase in grain yield, was achieved by going from T1 to T5 (Table 5). This indicates that there is significant yield advantage for wheat associated with the application of higher rates of blend fertilizer (i.e., up to T5). It is not necessarily recommendable, but the yields keep increasing, showing significant differences between T4 and T5. Therefore, under the current blend fertilizer approach, it can be concluded that application of 300 kg NPS + ZnB/ha could pay off, depending on the relative prices of wheat grain and fertilizer, and the production motives of the farm households. The same was noted for wheat on nitisols and andosols in the Ethiopian highlands [17].
Comparing the performance of the same amounts of the newly introduced blend (T3-NPS + ZnB) against conventionally used DAP fertilizer (T7) shows no significant differences for teff and wheat grain and biomass yields. This essentially means that there is no added yield advantage of incorporating S, Zn and B nutrients for both cereal crops, although B in soil samples coming out as significantly correlated to wheat grain and biomass (Table 4). When comparing NPS (T6) and NPS + ZnB (T2), the fertilizer without Zn and B performed better on teff while showing no significant difference for wheat. Adding the micronutrients also is at the expense of the N and P ratios in the blend, which may have a detrimental effect on yield, as N and P are the two most limiting nutrients for crop production in the Ethiopian highlands. This is in line with the finding of Brhane et al. [34], who suggested reduced N and P contents to include K containing a blend fertilizer from their wheat trial in the northern part of the country.
4. Conclusions
This study highlights how land use and soil management practices can affect the fertility status of soils, which in turn significantly affects crop yields and response to fertilizer application. Ethiopian vertisols exhibit a wide variation in properties that express themselves in the yields of teff and wheat. Differences in pedogenetic processes and biophysical factors are also important, as parent materials shape secondary clay minerals. Soil properties highly significantly affecting crop yields are soil pH, OC, AP, S, K, suggesting the importance of the proper management of these properties for higher yields on vertisols.
The results also indicate that fertilizer application alone is not a sufficient condition to attain higher crop yields. The current land use practices under the highland cereal crop livestock system (FS1 and FS2) have resulted in the depletion of OC and nutrient levels (N, S, and K), when compared to the enset–coffee–livestock system (FS3) and khat–sorghum system (FS4). This, in turn, significantly affected crop performance and response to fertilizer application. Under the current blend fertilizer application approach based on a NPS + ZnB blend, the study found that levels up to 100 kg/ha for teff and 300 kg/ha for wheat have benefits that could translate into recommendations if the prices of inputs and outputs allow. As S and Zn in the soils are not significantly correlated with teff and wheat yields, and the treatments that include NPS + ZnB do not outperform NPS or DAP alone, there is no justification for the adoption of a blend fertilizer. In the future, micronutrients may become limiting but right now, N and P are the most limiting nutrients. Given the diversity in land use practices, biophysical settings and soil fertility managements, a blanket approach has no place for improved crop yields. Therefore, we recommend, soil-, site- and crop-specific fertility recommendations. If management practices are matched with site conditions, vertisols can make enormous contributions to food production and grain self-sufficiency, which is an important policy agenda in Ethiopia at the moment.
Conceptualization, E.E. and G.K.B.; Methodology, E.E., G.K.B. and E.M.A.S.; Software, G.K.B.; Formal Analysis, E.E. and G.K.B.; Investigation, E.E., G.K.B. and E.M.A.S.; Resources, E.M.A.S.; Data Curation, E.E.; Writing—Original Draft Preparation, E.E. and G.K.B.; Writing—Review and Editing, E.M.A.S.; Visualization, E.E. and G.K.B.; Supervision, E.E.; Project Administration, E.E. and E.M.A.S.; Funding Acquisition, E.E. and E.M.A.S. All authors have read and agreed to the published version of the manuscript.
Not applicable.
Not applicable.
The data used in this paper can be accessed from the corresponding author up request.
The analytical data were generated as part of the BENEFIT-CASCAPE
The authors declare no conflict of interest.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Enlarge this image.Figure 1. Occurrence of vertisols and major land use systems in the Ethiopian highlands, and experimental fields.
Enlarge this image.Figure 2. Rainfall totals during the growing season at the experimental sites (2017, 2018).
Enlarge this image.Figure 3. Dung cake drying for sale as a household fuel in the vertisol area in the central highlands.
Enlarge this image.Figure 4. Teff yield and biomass response showing inverse relationship between grain and biomass for the two-growing season.
Biophysical properties of vertisol landscapes across land use systems in the Ethiopian highlands (location in
| Variable | Land Use System | ||||
|---|---|---|---|---|---|
| FS1 | FS2 | FS3 | FS4 | ||
| Elevation (m) | Min | 1514 | 1599 | 1908 | 1726 |
| Max | 2987 | 2954 | 2097 | 2339 | |
| Mean | 2222 | 2240 | 976 | 2099 | |
| Rainfall (mm) | Min | 450 | 800 | 1108 | 750 |
| Max | 800 | 1423 | 1301 | 975 | |
| Mean | 635 | 1187 | 1195 | 810 | |
| Tmin (°C) | Min | 11 | 10 | 15 | 10 |
| Max | 21 | 16 | 15 | 13 | |
| Mean | 15 | 13 | 15 | 11 | |
| Tmax (°C) | Min | 23 | 20 | 26 | 24 |
| Max | 28 | 27 | 26 | 27 | |
| Mean | 25 | 25 | 26 | 25 | |
| Slope (%) | Min | 1 | 1 | 1 | 1 |
| Max | 3 | 3 | 2 | 13 | |
| Mean | 2 | 2 | 1 | 6 | |
FS1, sorghum–teff–livestock mixed system; FS2, wheat–maize–teff–barley–livestock system; FS3, enset–coffee–cereal–livestock complex; and FS4, khat–sorghum–potato–livestock system; Tmin—the mean minimum temperature from all the stations within the same land use system; Tmax—the mean maximum temperature in degree Celsius.
Descriptive statistics of selected soil properties of vertisols across the four land use systems in the Ethiopian highlands (n = 45).
| Parameter | Farming System | Mean | SD | CV (%) | F | P | |||
|---|---|---|---|---|---|---|---|---|---|
| FS1 | FS2 | FS3 | FS4 | ||||||
| Sand (%) | 29 a,b | 33 a | 15 c | 21 b,c | 23 | 10.25 | 32.23 | 9.4715 | 0.001 |
| Silt (%) | 18 c | 23 a,b | 29 a | 21 b,c | 22 | 7.06 | 29.60 | 3.6378 | 0.020 |
| Clay (%) | 53 a,b | 44 c | 56 a,b | 58 a | 55 | 10.22 | 13.07 | 6.2852 | 0.001 |
| BD (g/cm3) | 1.44 a | 1.26 a,b | 1.05 c | 1.21 b,c | 1.22 | 0.11 | 7.73 | 5.7543 | 0.002 |
| pH-H2O | 7.07 a,b | 5.58 c | 6.53 b,c | 7.35 a | 6.63 | 0.88 | 10.96 | 9.3185 | 0.001 |
| OC (%) | 1.92 c | 1.54 c | 5.73 b | 13.13 a | 5.58 | 5.07 | 44.41 | 126.6 | 0.001 |
| TN (%) | 0.18 c | 0.17 c | 0.26 b | 1.42 a | 0.51 | 0.55 | 44.09 | 111.83 | 0.001 |
| AP (mg/kg) | 18 a | 7 c | 10 b | 8.49 b,c | 10.94 | 7.26 | 43.03 | 3.8122 | 0.016 |
| S (mg/kg) | 0.69 c | 0.75 b,c | 0.98 a,b | 1.08 a | 0.87 | 0.28 | 28.82 | 6.838 | 0.000 |
| Exchangeable bases and CEC (cmol(+)/kg) | |||||||||
| CEC | 48 b | 37 c | 53 a | 51 a,b | 49.23 | 9.11 | 17.27 | 4.7328 | 0.006 |
| Ca | 29 b | 29 b | 20 c | 33 a | 29.78 | 7.98 | 24.58 | 3.4907 | 0.024 |
| Mg | 10 b,c | 13 a | 7 c | 11 a,b | 10.69 | 2.77 | 21.92 | 9.0677 | 0.001 |
| Na | 0.98 a | 0.67 c | 0.91 b | 1.07 a | 0.90 | 0.42 | 46.56 | 2.222 | 0.100 |
| K | 0.41 b,a | 0.70 a | 0.59 a | 0.60 a | 0.56 | 0.31 | 54.95 | 0.5912 | 0.107 |
| Micro-nutrients (mg/kg) | |||||||||
| Fe | 14 d | 99 a | 32 b | 23 c | 31.55 | 33.16 | 70.52 | 19.713 | 0.001 |
| Mn | 10 c | 69 a | 26 b | 16 b,c | 23.01 | 23.50 | 66.40 | 18.05 | 0.001 |
| Zn | 1 c | 12 a | 2 b | 1 c | 2.06 | 4.53 | 146.60 | 26.581 | 0.001 |
| Cu | 3 a | 2 a | 2.21 a | 4 a | 2.80 | 1.98 | 64.88 | 2.193 | 0.103 |
| B | 0.38 a | 0.32 a | 0.16 b | 0.30 a | 0.29 | 0.16 | 16.55 | 17.021 | 0.000 |
FS1, sorghum–teff–livestock mixed system; FS2, wheat–maize–teff–barley–livestock system; FS3, enset–coffee–cereal–livestock complex; FS4, khat–sorghum–potato–livestock system. SD, standard deviation of the mean; CV, coefficient of variation; BD, bulk density. Location means are compared across rows, and means within a row that have similar letters are not significantly different at p < 0.05.
Variation in teff and wheat biomass and grain yield on vertisols across locations.
| Location/Crop | Plant Height (cm) | Biomass Yield (t/ha) | Grain Yield (t/ha) |
|---|---|---|---|
| Teff | |||
| Burie (FS2) | 110.98 b | 10.38 b | 1.72 b |
| Becho (FS2) | 90.86 c | 9.17 b | 2.15 a |
| Enamore (FS3) | 131.89 a | 12.47 a | 1.34 c |
| Mean | 111.24 | 10.67 | 1.74 |
| F | 126.09 | 18.71 | 69.53 |
| P | <0.001 | 0.001 | 0.001 |
| Wheat | |||
| Endamohoni (FS1) | 133.15 a | 14.03 a | 5.78 a |
| Girar Jarso (FS2) | 118.09 b | 5.26 b | 1.95 b |
| Mean | 125.62 | 9.64 | 3.87 |
| F | 12.142 | 689.979 | 792.771 |
| p | 0.002 (**) | 0.001 (***) | 0.001 (***) |
Means within column followed by the same letter are not significantly different from each other at p < 0.05 (*); p < 0.01 (**); and p < 0.001 (***) levels of significance.
The correlation between selected soil properties on teff and wheat grain and biomass yield.
| GY | BMY | pH | OC | TN | AP | S | K | Fe | Mn | Zn | Cu | B | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Relation for teff | |||||||||||||
| GY | 1.00 | ||||||||||||
| BMY | −0.11 | 1.00 | |||||||||||
| pH | 0.32 * | −0.44 ** | 1.00 | ||||||||||
| OC | −0.45 ** | 0.37 ** | −0.64 | 1.00 | |||||||||
| TN | 0.03 | 0.25 * | −0.12 | 0.02 | 1.00 | ||||||||
| P | 0.05 | 0.07 | 0.15 | −0.05 | −0.01 | 1.00 | |||||||
| S | −0.05 | 0.32 ** | −0.72 *** | 0.54 ** | 0.03 | −0.24 * | 1.00 | ||||||
| K | 0.44 ** | −0.42 ** | 0.86 *** | −0.85 | −0.03 | 0.24 * | −0.68 | 1.00 | |||||
| Fe | −0.52 ** | 0.43 ** | −0.73 *** | 0.80 *** | 0.10 | −0.07 | 0.52 ** | −0.83 *** | 1.00 | ||||
| Mn | −0.44 ** | 0.47 ** | −0.75 *** | 0.87 *** | 0.06 | −0.08 | 0.53 ** | −0.88 *** | 0.78 *** | 1.00 | |||
| Zn | −0.10 | 0.08 | −0.33 * | 0.34 * | 0.01 | 0.28 * | 0.39 ** | −0.35 * | 0.40 * | 0.30 * | 1.00 | ||
| Cu | 0.00 | 0.29 * | −0.79 *** | 0.35 * | 0.19 | −0.37 * | 0.78 *** | −0.44 * | 0.45 * | 0.47 ** | 0.25 * | 1.00 | |
| B | 0.23 * | −0.39 * | 0.54 ** | −0.34 * | −0.15 | 0.38 * | −0.52 ** | 0.41 * | −0.42 * | −0.42 ** | −0.24 * | −0.60 *** | 1 |
| Relation for wheat | |||||||||||||
| GY | 1.00 | ||||||||||||
| BMY | 0.99 *** | 1.00 | |||||||||||
| pH | 0.56 ** | 0.61 ** | 1.00 | ||||||||||
| OC | 0.86 *** | 0.87 *** | 0.54 ** | 1.00 | |||||||||
| TN | 0.68 ** | 0.68 ** | 0.32 * | 0.83 ** | 1.00 | ||||||||
| AP | 0.53 ** | 0.53 ** | 0.30 * | 0.40 ** | 0.21 | 1.00 | |||||||
| S | −0.43 * | −0.41 ** | −0.21 | −0.34 * | −0.26 | −0.22 | 1.00 | ||||||
| K | −0.84 *** | −0.84 ** | −0.47 | −0.88 | −0.77 | −0.36 | 0.58 | 1.00 | |||||
| Fe | 0.37 * | 0.37 * | 0.25 * | 0.40 ** | 0.42 * | 0.47 * | −0.14 | −0.26 * | 1.00 | ||||
| Mn | 0.15 | 0.15 | 0.11 | 0.17 | 0.28 | 0.39 * | −0.09 | −0.06 | 0.92 *** | 1.00 | |||
| Zn | 0.12 | 0.16 | −0.13 | 0.06 | −0.03 | 0.21 | 0.03 | −0.11 | 0.10 | 0.16 | 1.00 | ||
| Cu | 0.78 *** | 0.79 *** | 0.61 ** | 0.88 ** | 0.74 ** | 0.42 * | −0.37 * | −0.80 | 0.64 ** | 0.45 * | 0.09 | 1.00 | |
| B | 0.60 ** | 0.62 ** | 0.23 * | 0.71 ** | 0.59 ** | 0.26 | −0.36 * | −0.66 | 0.24 | 0.09 | 0.17 | 0.64 ** | 1 |
GY = grain yield; BMY = biomass yield; the asterisk indicates levels of significance p < 0.05 (*); p < 0.01 (**); & p < 0.001 (***).
Effect of fertilizer treatments on teff and wheat yield on vertisols.
| Fertilizer Treatment | Teff Yield (t/ha) | Wheat Yield (t/ha) | |||
|---|---|---|---|---|---|
| Biomass | Grain | Biomass | Grain | ||
| 50 NPSZnB (T1) | 8.83 b | 1.45 c | 7.26 d | 2.65 c | |
| 100 NPSZnB (T2) | 10.50 a,b | 1.62 c,b | 9.61 b,c | 3.95 b | |
| 150 NPSZnB (T3) | 10.92 a,b | 1.75 a,b,c | 9.44 b,c | 3.66 b | |
| 200 NPSZnB (T4) | 11.31 a,b | 1.95 a | 11.01 a,b | 4.19 b | |
| 300 NPSZnB (T5) | 12.26 a | 1.93 b,a | 11.98 a | 5.03 a | |
| 100 NPS (T6) | 10.27 a,b | 1.82 b,a | 8.94 c,d | 3.64 b | |
| 150 DAP (T7) | 10.64 a,b | 1.64 a,b,c | 9.27 b,c | 3.96 b | |
| Mean | 10.68 | 1.740 | 9.648 | 3.870 | |
| F | 3.15 | 5.98 | 11.721 | 15.607 | |
| P | 0.008 (**) | 0.001 (***) | 0.001 (***) | 0.001 (***) | |
Means within column followed by the same letter are not significantly different from each other at p < 0.05 (*); p < 0.01 (**); and p < 0.001 (***) levels of significance.
References
1. Santra, P.; Kumar, M.; Panwar, N.R.; Das, B. Digital Soil Mapping and Best Management of Soil Resources: A Brief Discussion with Few Case Studies. Adaptive Soil Management: From Theory to Practices; Rakshit, A.; Abhilash, P.C.; Singh, H.B.; Ghosh, S. Springer Nature Singapore Pte Ltd.: Singapore, 2017; pp. 3-39.
2. Somasundaram, J.; Lal, R.; Sinha, N.K.; Dalal, R.; Chitralekha, A.; Chaudhary, R.S.; Patra, A.K. Cracks and Potholes in Vertisols: Characteristics, Occurrence, and Management. Adv. Agron.; 2018; 149, pp. 93-159.
3. Kovda, I. Vertisols: Extreme features and extreme environment. Geoderma Reg.; 2020; 22, e00312. [DOI: https://dx.doi.org/10.1016/j.geodrs.2020.e00312]
4. Elias, E. Soils of the Ethioian Highlands: Geomorphology and Properties; Capacity Building for Scaling Up of Evidence-Based Best Practices for Increased Agricultural Production in Ethiopia (CASCAPE) Wageningen University and Research: Wageningen, The Netherlands, 2016; 385.
5. Dinssa, B.; Elias, E. Characterization and classification of soils of Bako Tibe District, West Shewa, Ethiopia. Heliyon; 2021; 7, e08279. [DOI: https://dx.doi.org/10.1016/j.heliyon.2021.e08279] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34765792]
6. Viaud, V.; Santillàn-Carvantes, P.; Akkal-Corfini, N.; Le Guillou, C.; Prévost-Bouré, N.; Ranjard, L.; Menasseri-Aubry, S. Landscape-scale analysis of cropping system effects on soil quality in a context of crop-livestock farming. Agric. Ecosyst. Environ.; 2018; 265, pp. 166-177. [DOI: https://dx.doi.org/10.1016/j.agee.2018.06.018]
7. Rossiter, D.G.; Bouma, J. A new look at soil phenoforms—Definition, identification, mapping. Geoderma; 2018; 314, pp. 113-121. [DOI: https://dx.doi.org/10.1016/j.geoderma.2017.11.002]
8. Zewdie, E. Properties of Major Agricultural Soils of Ethiopia; LAP Lambert Academic Publishing: Saarbrucken, Germany, 2013; 281.
9. Shabtai, I.A.; Shenker, M.; Edeto, W.L.; Warburg, A.; Ben-Hur, M. Effects of land use on structure and hydraulic properties of Vertisols containing a sodic horizon in northern Ethiopia. Soil Tillage Res.; 2014; 136, pp. 19-27. [DOI: https://dx.doi.org/10.1016/j.still.2013.09.007]
10. Erkossa, T.; Haileslassie, A.; MacAlister, C. Enhancing farming system water productivity through alternative land use and water management in vertisol areas of Ethiopian Blue Nile Basin (Abay). Agric. Water Manag.; 2014; 132, pp. 120-128. [DOI: https://dx.doi.org/10.1016/j.agwat.2013.10.007]
11. Hailu, H.; Mamo, T.; Keskinen, R.; Karltun, E.; Gebrekidan, H.; Bekele, T. Soil fertility status and wheat nutrient content in Vertisol cropping systems of central highlands of Ethiopia. Agric. Food Secur.; 2015; 4, 19. [DOI: https://dx.doi.org/10.1186/s40066-015-0038-0]
12. Pierre, T.J.; Primus, A.T.; Simon, B.D.; Philemon, Z.Z.; Hamadjida, G.; Monique, A.; Pierre, N.J.; Lucien, B.D. Characteristics, classification, and genesis of Vertisols under seasonally contrasted climate in the Lake Chad Basin, Central Africa. J. Afr. Earth Sci.; 2018; 150, pp. 176-193. [DOI: https://dx.doi.org/10.1016/j.jafrearsci.2018.11.003]
13. Astatke, A.; Jabbar, M.A. Low-cost Animal-drawn Implements for Vertisol Management and Strategies for Land-use Intensification. The Sustainable Management of Vertisols; Syers, J.K.; Penning de Vries, F.W.T.; Nyamudeza, P. CABI Publisher: London, UK, 2001.
14. Jabbar, M.A.; Mamo, T.; Mohamed Saleem, M.A. From Plot to Watershed Management: Experience in Farmer Participatory Vertisol Technology Generation and Adoption in Highland Ethiopia. The Sustainable Management of Vertisols; Syers, J.K.; Penning de Vries, F.W.T.; Nyamudeza, P. CABI Publisher: London, UK, 2001.
15. CDCR. Poverty and Hunger Strategic Review. Ethiopia Roadmap to Achieve Zero Poverty and Hunger; Centre for Dialogue, Research and Cooperation (CDRC): Addis Ababa, Ethiopia, 2019.
16. Abdulkadir, B.; Kassa, S.; Desalegn, T.; Tadesse, K.; Haileselassie, M.; Fana, G.; Abera, T.; Amede, T.; Tibebe, D. Crop response to fertilizer application in Ethiopia: A review. Proceedings of the Crop Response to Fertilizer Rapplication; Addis Ababa, Ethiopia, 1–2 December 2016.
17. Elias, E.; Okoth, P.F.; Smaling, E.M.A. Explaining bread wheat (Triticum aestivum) yield differences by soil properties and fertilizer rates in the highlands of Ethiopia. Geoderma; 2019; 339, pp. 126-133. [DOI: https://dx.doi.org/10.1016/j.geoderma.2018.12.020]
18. Leenaars, J.G.B.; Elias, E.; Wösten, H.; Ruiperez, M.; Kempen, B.; Ali, A.; Brouwer, F. Major Soil-Landscape Resources of the Cascape Intervention Woredas, Ethiopia: Soil Information in Support to Scaling Up of Evidence-Based Best Practices in Agricultural Production (with Dataset); ISRIC-ALTERA, WUR: Wageningen, The Netherlands, 2016.
19. Leenaars, J.G.B.; Elias, E.; Wösten, J.H.M.; Ruiperez-González, M.; Kempen, B. Mapping the major soil-landscape resources of the Ethiopian Highlands using random forest. Geoderma; 2020; 361, 114067. [DOI: https://dx.doi.org/10.1016/j.geoderma.2019.114067]
20. Van Reeuwijk, L.P. Procedures for Soil Analysis; 6th ed. International Soil Reference and Information Centre (ISRIC): Wageningen, The Netherlands, 2006.
21. Bremner, J.M.; Mulvaney, C.S. Total Nitrogen. Methods of Soil Analysis II. Chemical and Microbiological Properties; Page, A.L.; Miller, R.H.; Keeney, D.R. Soil Science Society of America: Madison, WI, USA, 1982; pp. 595-642.
22. Mylavarapu, R.S.; Sanchez, J.F.; Nguyen, J.H.; Bartos, J.M. Evaluation of Mehlich-1 and Mehlich-3 extraction procedures for plant nutrients in acid mineral soils of Florida. Commun. Soil Sci. Plant Anal.; 2002; 33, pp. 807-820. [DOI: https://dx.doi.org/10.1081/CSS-120003067]
23. Tan, K. Soil Smapling, Preparation, and Analysis; Marcel Dekker, Inc.: New York, NY, USA, 1996.
24. R Core Team. R: A Language and Environment for Statistical Computing; R Foundation for Statistical Computing: Vienna, Austria, 2019.
25. Fox, J.; Weisberg, S. An R Companion to Applied Regression; 3rd ed. Sage: Thousand Oaks, CA, USA, 2019.
26. Eswaran, H. Taxonomy and management related properties of the red soils of Africa. Proceedings of the International Symposium on the Red Soils of East and Southern Africa; Harare, Zimbabwe, 24–27 February 1986.
27. Tadesse, K.A. Effects of Traditional Practice of Soil Burning (Guie) on Soil Chemical Properties at Sheno Areas of North Shoa, Oromia Region, Ethiopia. J. Plant Sci.; 2015; 3, pp. 342-348.
28. Negasa, T.; Ketema, H.; Legesse, A.; Sisay, M.; Temesgen, H. Variation in soil properties under different land use types managed by smallholder farmers along the toposequence in southern Ethiopia. Geoderma; 2017; 290, pp. 40-50. [DOI: https://dx.doi.org/10.1016/j.geoderma.2016.11.021]
29. Elias, E. Characteristics of Nitisol profiles as affected by land use type and slope class in some Ethiopian highlands. Environ. Syst. Res.; 2017; 6, 20. [DOI: https://dx.doi.org/10.1186/s40068-017-0097-2]
30. Elias, E. Selected chemical properties of agricultural soils in the Ethiopian highlands: A rapid assessment. S. Afr. J. Plant Soil; 2018; 36, pp. 153-156. [DOI: https://dx.doi.org/10.1080/02571862.2018.1506829]
31. Elmobarak, A.A. Status, priorities and needs for sustainable soil management in Sudan. Proceedings of the Global Soil Partnership in East and Southern Africa; Nairobi, Kenya, 25–27 March 2013.
32. Moussadek, R.; Laghrour, M.; Mrabet, R.; Van Ranst, E.; Badraoui, M.; Mekkaoui, M. Morocco’s Vertisol Characterization. J. Mater. Environ. Sci.; 2017; 8, pp. 3932-3942.
33. Fageria, N.K.; Baligar, V.C.; Jones, C.A. Growth and Mineral Nutrition of Field Crops; 3rd ed. Taylor and Francis Group, LLC: New York, NY, USA, 2011.
34. Brhane, H.; Mamo, T.; Teka, K. Optimum potassium fertilization level for growth, yield and nutrient uptake of wheat (Triticum aestivum) in Vertisols of Northern Ethiopia. Cogent Food Agric.; 2017; 3, 1347022. [DOI: https://dx.doi.org/10.1080/23311932.2017.1347022]
© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Details
; Gizachew Kebede Biratu 2
; Smaling, Eric M A 3
1 Centre for Environmental Science, College of Natural and Computational Sciences, Addis Ababa University, Addis Ababa P.O. Box 1176, Ethiopia;
2 Department of Natural Resource Management, School of Natural Resource, Guder Mamo Mezemir Campus, Ambo University, Ambo P.O. Box 19, Ethiopia
3 Wageningen Environmental Research, Wageningen University & Research, 6708 Wageningen, The Netherlands;