1 Introduction

Ethiopia has the largest livestock population in Africa. According to the Central Statistical Agency [1], the country has approximately 70 million head of cattle, 42.9 million head of sheep, 52.5 million head of goats, 2.15 million head of horses, 10.80 million head of donkeys, 0.38 million head of mules, 8.1 million head of camels, and 56.87 million head of poultry at the country level. Despite having one of the largest livestock populations, its productivity has remained very low. This is mainly attributed to feed shortages both in quality and quantity [2].

Livestock production contributes approximately 17–25.3% of gross domestic product (GDP), 39–49% of agricultural GDP and more than 50% of household income in Ethiopia [3]. In addition, livestock supplies meat, milk and eggs as sources of protein and significantly contributes to the sustainability of crop production through the provision of draught power and manure for soil fertilization, transportation, income, employment, manure for soil fertility management, and security to reduce the risk of crop failure, live banks, wealth storage, and social prestige for rural farmers [4].

Despite the large population, animal productivity in Ethiopia is poor, even below the values recorded for the majority of the countries in eastern and sub-Saharan Africa [5]. The major constraints are low-productivity feed scarcity in terms of quality and quantity, low genetic potential of indigenous livestock, a high incidence of disease and parasites, traditional management practices, poor access to extension and credit services and a lack of knowledge to improve animal performance [6,7,8]. Of these, poor-quality feed resources were identified as the greatest constraint limiting livestock productivity [8].

Natural pasture, crop residues, hay, agro-industrial byproducts, improved feed and other products, such as animal byproducts and vegetable and fruit wastes, are livestock feed resources available in Ethiopia [9]. Of these, natural pasture and crop residues are major feed resources but are characterized by low digestibility and crude protein content, which adversely affects livestock productivity 3 (Shapiro et al. [3]). In southern Ethiopia [10], reported that the CP contents of grass species and crop residues ranged from 1.42–18.95% to 2.01–8.97% DM, respectively. In terms of quality, cereal crop residues are generally of low nutritive value because of their relatively low CP ranging from 2.39 ± 0.16% to 7.20 ± 0.09% and high NDF ranging from 72.59 ± 2.26 to 78.19 ± 0.77% in western Ethiopia [11]. Leng [12] defined low-quality forage as those with a CP value < 8% and suggested the supplementation of such forages with appropriate nutrients to achieve high levels of animal production.

Ref [13] reported that the quality of feeds reveals the nutrient (chemical) composition, palatability and intake, digestibility, antinutritional factors and animal production performance. Most of the forages from natural pastures and crop residues contain CP below 7% and NDF above 55% DM in Ethiopia [14], both of which indicate poor nutritive value incapable of meeting rumen microbial requirements, particularly with regard to CP content (Van Soest, [15]). Ref [12] also indicated that low-quality forages, such as those with CP content less than 8% DM CP content, adversely affect rumen microbial activity [15]. It has been stated that a CP content of approximately 15% DM for high milk production (> 15 kg/cow/day) and 8–13% for moderate milk production (10–15 kg/cow/day) are required by dairy cows [16, 17].

The total NDF content of forage is a dominant factor in determining forage quality [18]. The NDF content of feeds above 60% and 50–60% DM are classified as poor and moderate quality feed, respectively 16 (Van Soest, [19]). A greater amount of forage ADF results in reduced digestibility of dry matter as a consequence of increased lignification of cellulose in plants [20]. Kellems and Church [21] categorized roughages with less than 40% ADF as high quality and those with more than 40% ADF as low quality. The lignin content of feeds and forages affects the digestibility of forage more than any other chemical component of feeds [19, 22]. According to Van Soest [19], a lignin content above 6% DM negatively affects the digestibility of forage.

Deficiency of feed in terms of quantity and quality negatively affects the productive and reproductive performance of grazing livestock [23]. According to the NRC [24], optimum productive and reproductive performance of livestock can be achieved only when the animals are fed the required quantity of feedstuffs containing all the nutrients in the proper amount. Precise information on feed composition with respect to proximate composition and fiber fractions is essential for assessing the nutritional status of feeds and fodders and of the animals to which these feeds and fodders are fed [25]. However, the nutritional value of feed resources varies depending on environmental factors such as altitude (agro-ecological zone, rainfall, soil type, cropping intensity, grazing land management and genetic characteristics specific to plant species [26].

In the present study area, natural pasture, crop residues, indigenous fodder trees and shrubs, and, to a lesser extent, improved forages are the most important feed resources for livestock. However, there is no information on their chemical composition. This lack of information could limit the formulation of balanced rations for better utilization of available feed resources to improve livestock productivity. Thus, the results of this study will help to fill this information gap and provide the basis for generating empirical evidence to improve feed quality to improve livestock productivity. With this insight, the objective of this study was to evaluate and compare the chemical composition of major feed resources obtained from three agro-ecological zones of the Gera district, southwest Ethiopia.

2 Materials and methods

2.1 Description of the study area

This study was conducted in the Gera district of the Jimma zone, Oromia Regional State, southwest Ethiopia. A detailed description of the study area can be found in a previously published companion study [27].

2.2 Feed sampling

Representative samples of the most commonly available livestock feedstuffs including three natural pasture species (Cynodon dactylon, Pavoniaschimperiana Hochst and Rhynchosia ferruginea), three indigenous fodder trees and shrubs (Erythrina Abyssinica, Vernonia amygdalina, and Maytenus undat), two cultivated forages (Pennisetum purpureum and Pennisetum pedicellatum) and five crop residues (Hordeum vulgare, Zea mays stover, Sorghum bicolor stover, Triticum aestivum straw and Eragrostis tef straw) were randomly collected from the HL, ML and LL AEZs of the Gera district following appropriate procedures. Samples of natural pasture were collected using a 0.5 × 0.5 m quadrate from 20 major grazing land sites in each AEZ during October to December (early dry season),and then were bulked together on AEZ basis and then thoroughly mixed and sub-sumpled.The quadrate was thrown on the selected sites and harvested the whole plant at grazing height of about 5 cm above the ground to mimic the way ruminants graze. For indigenous fodder trees and shrubs, edible green leaves and twigs were hand harvested from all three directions in the lower, middle, and top portions of the canopy in each agro-ecological zone. All samples of crop residues stems (including leaves, stems or stalks) were collected from the crop fields after crops had been harvested and/or threshed under traditional practices. The crops were harvested after they were dry. To produce one representative composite sample for each species, the samples were bulked together and mixed thoroughly separately in each AEZ. The fresh samples of native grasses, improved forages and IFTSs were weighed, sun-dried and stored in polythene bags before being transported to the laboratory. Chemical analyses of feed samples were performed at Animal Nutrition Laboratory of the Department of Animal Science, College of Agriculture and Veterinary Medicine, Jimma University.

2.3 Laboratory analysis

In the laboratory, all samples were oven dried at 65 °C for 72 h were used for laboratory analysis to determine chemical composition. The dried samples were then milled to pass through a 1 mm mesh sieve. The samples were analyzed on a % DM basis for ash, CP, EE and CF by the procedures of the [28]. Total ash content was determined by oven drying the samples at 105 °C overnight and by combusting the samples in a muffle furnace at 550 °C for 6 h. Nitrogen (N) content was determined following the micro-Kjeldahl digestion, distillation and titration procedures [29] and the CP content was estimated by multiplying the N content by 6.25. The structural plant constituents (NDF, ADF and ADL) were determined according to Van Soest et al. [30]. The samples were analyzed in duplicate.

2.4 Statistical analysis

The data were subjected to analyses of variance (ANOVAs) using the Statistical Package for Social Sciences (SPSS) Program Version 20.0. The differences between means were separated by Duncan’s multiple range test using the IBM SPSS Statistics Programme, Version 20.0. In addition, T-test was used to see significance difference between the various parameters of the two improved forage species. Differences were considered significant when P < 0.05. The results are presented as the means ± standard errors (SEs).

The statistical model used for data analysis was as follows:

$${\text{Yij }} = \, \mu \, + \alpha {\text{i }} + \sum {{\text{ij}}}$$

where Yij is the response of the parameter/variable investigation; DM, Ash, CP, CF, EE, NDF, ADF and ADL.

μ = overall mean.

αI = the effect of ith location/agro-ecology (I = HL, ML, LL).

Σij = random error.

3 Results and discussion

3.1 Chemical compositions of natural pasture species

Table 1 shows the chemical composition (% DM) of the natural pastures in the three agro-ecological zones of the study area. The results showed that the mean DM, ash, EE, CP, NDF, ADF, and CF of natural pastures ranged from 89.82 ± 0.29 in Rhynchosia ferruginea A. Rich at HL to 93.58 ± 0.26 in C. dactylon at LL, from 7.95 ± 0.39 in C. dactylon at LL to 14.83 ± 0.08 in Pavonia Schimperiana Hochst at HL, from 1.13 ± 0.09 in Pavonia Schimperiana Hochst at ML to 1.62 ± 0.12 in C. dactylon at HL, 7.12 ± 0.72 in Pavonia Schimperiana Hochst at ML to 11.16 ± 0.44 in C. dactylon at HL, 61.12 ± 0.64 in Pavonia Schimperiana Hochst at HL to 69.34 ± 0.44 in C. dactylon at LL, 37.41 ± 0.44 in Rhynchosia ferruginea A. Rich at HL to 47.68 ± 0.46 in C. dactylon at LL, and 6.59 ± 0.

Table 1 Chemical composition (% DM) of natural pastures in three AEZs of Gera District, southwest Ethiopia
Full size table

The DM, ash, EE, CP, NDF, ADF, ADL and CF values of Cynadon dactylon ranged from 90.95 ± 0.46 to 93.58 ± 0.26, 9.27 ± 0.31 to 9.63 ± 0.68, 1.35 ± 0.15 to 1.62 ± 0.12, 7.77 ± 0.14 to 11.16 ± 0.44, 61.16 ± 0.86 to 69.34 ± 0.44, 38.24 ± 0.86 to 47.68 ± 0.46, 6.76 ± 0.73 to 11.56 ± 0.92, and 34.52 ± 0.38 to 43.59 ± 0.88, respectively. Variation in AEZ significantly (p < 0.05) influenced the chemical composition parameters of Cynadon dactylon. The DM, NDF, AD, AD, and CF contents of C. dactylon were significantly (p < 0.05) greater at LL compared to other AEZS, whereas CP content was greater (p < 0.05) at HL than ML and LL. AEZ had no effect on the ash or EE content of C. dactylon (p > 0.05). These variations might be due to climatic conditions, soil fertility, and harvesting stage.

Pavonia schimperiana from HL exhibited significantly (p < 0.05) greater ash values than those from ML and LL, whereas CP and EE were greater at LL, and NDF was greater at ML than at other AEZs. The DM, ADF and ADL contents of Rhynchosia ferruginea A. Rich in ML were significantly (p < 0.05) greater than those in HL and LL. The variations in the chemical composition traits could be due to species, soil fertility, agro-ecology and climate. The other chemical composition parameters did not vary among the agro-ecological zones. The variations observed in the various chemical composition parameters among the different species in the AEZs might be due to differences in climatic conditions, soil fertility and stage of maturity.

The DM values of the natural pasture species observed in this study are greater than the recommended range of 70–80% and may limit feed intake by livestock [15].

The CP content of the natural pasture obtained in this study was greater than the minimum of 7–8% DM for optimum rumen microbial function and maintenance requirements of ruminants (NRC, [31]). However, our results are lower than the minimum recommended values of 12% for lactation and 11.3% DM for growth in ruminants [16], indicating the need for supplementation with high-protein feeds. The voluntary intake of ruminants decreases when the CP level is below 6–7% DM [16, 32].

The EE content of natural pasture species observed in the present study was lower than the minimum recommended value of 5% DM, indicating a lower energy level for the animal [33]. Fats, as livestock feed, function much like carbohydrates in that they serve as a source of heat and energy and for the formation of fat due to the larger proportion of carbon and hydrogen.

The overall mean NDF content of the natural pasture recorded in this study was greater than the critical value of 60% [34], which may have resulted in decreased voluntary feed intake and feed conversion efficiency and increased rumination time. Roughage feeds with NDF contents less than 45, 45–65, and greater than 65% are considered high-, medium-, and low-quality, respectively [35]. It has been reported that 36% NDF is ideal for forage for domestic animals, but greater than 36% NDF limits of intake due to rumen fill, and less than 36% NDF results in insufficient fiber for rumen scratch factor and proper rumen function [31]. The higher NDF level of the natural pasture in this study might be due to its high maturity, which provided a chance for fiber accumulation in plant tissues.

In this study, the ADF content in natural pastures ranged from 37.41 ± 0.44 in Rhynchosia ferruginea A. Rich at HL to 47.68 ± 0.46 in Cynodon dactylon at LL. These results are higher than the recommended level of 18–20% DM [36] and the range of 17–21%, which is usually recommended for rumen stability [31]. ADF contents greater than 40% are considered low quality, whereas those less than 40% are considered high quality [21]. Based on this classification, the ADF contents of the natural pastures observed in this study were classified as both good and poor quality.

The ADL content of the natural pasture in this study ranged from 6.59 ± 0.73 in Rhynchosia ferruginea A. Rich at HL to 11.56 ± 0.92 in Cynodon dactylon at LL. Of the components of the cell wall, lignin is considered the main factor limiting feed intake, fiber degradation in the rumen, the rate of organic matter fermentation, the number of microbial cells produced per unit of fermented organic matter, and the proportion of propionate to acetate in the products of fermentation [15, 37]. The percentage of fiber that is digested may be less than 60% in feed that contains 10% DM of lignin [38]. Generally, the variations in the chemical composition of natural pastures between the present study and the literature might be due to environmental conditions, climatic conditions, plant type and species, soil fertility, weather conditions during growth, and stage of maturity variation.

3.2 Chemical composition of indigenous fodder trees and shrubs

Table 2 shows the chemical composition (on a % DM basis) of IFTSs in three agro-ecological zones of the study area. The mean DM, EE, NDF, ADF, ADL and CF contents of Vernonia amygdalina were significantly (P < 0.05) greater in LL AEZs than in ML and LL AEZs, while the ash content was significantly (P < 0.05) greater in ML AEZs than in HL and LL AEZs.

Table 2 Mean (± SE) chemical composition (% DM) of indigenous fodder trees and shrubs in different AEZs of the Gera district, Ethiopia
Full size table

All the chemical composition variables of Erythrina abyssinica were affected (P < 0.05) by the AEZ, except for DM and CF. The ash and CP contents were significantly (P < 0.05) greater in the HL AEZs than in the ML and LL AEZs, whereas NDF, ADF and ADL were significantly (P < 0.05) greater in the LL AEZ than in the HL and ML AEZ. The results revealed that AEZ had a significant effect (P < 0.05) on the chemical composition of Maytenus undata except for the EE. The NDF, ADF, ADL and CF contents of Maytenus undata were significantly (P < 0.05) greater in the ML than in the HL and LL AEZs. However, the DM and CP contents were significantly (P < 0.05) greater in LL than in ML and HL AEZ.

The DM content of IFTSs differed significantly (p < 0.05) across the AEZs. The mean DM content in the ML AEZs ranged from 88.49 ± 0.26 in Vernonia amygdalina to 94.07 ± 0.37 in Erythrina Abyssinica. The DM content of IFTSs observed in this study is in agreement with previous studies [10, 39] in which the DM content ranged from 88 to 94.55%. Andualem et al. [40] reported a DM value of 57.08% for browse species, which is much lower than the mean DM content obtained in this study. The difference may be due to variations in altitude, species, and climate.

The mean ash content of IFTSs ranged from 7.87 ± 0.72 in Maytenus undata at LL to 13.00 ± 0.38 in Maytenus undata at HL. The ash content in IFTSs recorded in this study is almost consistent with the findings of [10], who reported values ranging from 8.07 to 13.39%. Andualem et al. [40] reported an ash content of 5% DM, which is lower than the ash value obtained in this study. These differences could be attributed to and maturity stage. The mean EE content of IFTSs in this study varied from 1.78 ± 0.12% in Vernonia amygdalina at ML to 2.69 ± 0.14% DM in Erythrina abyssinica at ML.

Tamene et al. [11] reported lower EE content of IFTSs (1.64 ± 0.04%) in the LL and higher (4.10 ± 0.03%) in the ML AEZ compared to the findings of this study. This variation might be due to species of IFTSs, maturity and soil fertility status.

In this study, the average CP content of IFTSs ranged from 8.30 ± 0.44 in Erythrina abyssinica at LL to 11.85 ± 0.85 in Vernonia amygdalina at HL. The CP content in IFTSs reported in this study is greater than the recommended minimum threshold of 7% CP necessary for ruminant feed intake and optimum rumen microbial functions (Van Soest, [15]). The CP contents of IFTSs obtained in the current study is lower than the range of earlier studies [39, 40], which reported CP values ranging between 8.9 and 20.9% DM for indigenous browse species. Intake declines sharply when forage contains < 7% CP [41]. According to Mekonnen et al. [42], browse species can be used as good protein supplements for low-quality basal diets, especially during the dry season when the quality and quantity of green herbages are limited. Based on the findings of the present study, the CP content of IFTSs was greater than the acceptable threshold (7% DM CP). Thus, they have the potential to supplement low-quality crop residues and natural pastures, especially during the dry season.

The NDF content of IFTSs in this study ranged between 54.43 ± 1.28% in Erythrina abyssinica at HL and 66.67 ± 0.66% DM in Vernonia amygdalina at LL. The NDF content of all IFTS species was greater than 55% DM, the level above which voluntary feed intake is limited [42]. A range of 60–65% DM NDF is suggested as the limit above which the intake of tropical feeds by ruminants is limited [30]. Vernonia amygdalina was the most fibrous, with the highest overall NDF and ADF contents of 62.42 ± 1.72% and 47.81 ± 1.29% DM, respectively. An NDF range of 35–40% has been recommended by [43, 44] to be within the normal range for nutritious fodders.

The mean ADF content of IFTSs varied from 34.65 ± 1 in Maytenus undata al LL to 47.81 ± 1.29 in Vernonia amygdalina at the LL AEZ. The ADF contents of IFTSs recorded in this study are 15.55% greater than the DM reported for browse species by [40]. The ADF content of all IFTSs included in this study was greater than the reported range of 17–21%, which is usually recommended for rumen stability [31].

The mean ADL content of the IFTSs varied from 6.52 ± 0.74 in Maytenus undata to 11.65 ± 0.89 in Erythrina abyssinica in the LL AEZ. Khanal and Subba [45] reported that a high ADL content can limit the voluntary feed intake, digestibility, and nutrient utilization of ruminant animals. Generally, the variation in the chemical composition of IFTSs recorded in the present study and in the literature might be due to differences in the agro-ecological zone, climatic conditions, plant species, soil fertility conditions, weather conditions during growth, and stage of maturity.

3.3 Chemical composition of improved forages

Table 3 shows the chemical compositions (%DM basis) of the cultivated forage species in the three agro-ecological zones of the study area. The DM, Ash, EE, CP, NDF, ADF, ADL and CF values of Pennisetum purpureum were 89.17%, 10.86%, 1.6%, 6.9%, 66.69%, 34.57%, 12.22% and 45.92%, respectively.

Table 3 Chemical composition (% DM) of improved forages in three AEZs in the Gera district, southwest Ethiopia
Full size table

According to McDonald et al. [46], the DM content of fodder and formulated feeds influences the availability of nutrients and microbial activity. The DM content of Pennisetum purpureum (89.17%) recorded in the present study is in line with the results of [47], who reported 89.5% DM. However, this value is lower than the value (91%) reported by [48]. The ash content of Pennisetum purpureum (10.86%) obtained in this study is consistent with the observation of [48], who reported a value of 10.98%. The EE content of Pennisetum purpureum was very low (1.6% DM). It has been reported that EE contents of feeds above 7% DM limit the amount of feed that livestock consume [49].

The CP content of Pennisetum purpureum (6.9%) observed in the present study is higher than the reported value of 5.58% [50] but lower than the recommended value of 7 to 8% DM CP, which is the lowest amount of CP required for microbial growth in the rumen [15, 51]. The NDF in Pennisetum purpureum (66.69%) obtained in the current study is in agreement with the value of 67.11% reported by [47]. The percentage of ADF in Pennisetum purpureum (34.57%) recorded in the present study is lower than the 47.45% reported by [50]. The ADF value of Pennisetum purpureum is greater than the reported range of 17–21% recommended for rumen stability [31].

The DM, Ash, EE, CP, NDF, ADF, ADL and CF contents of Pennisetum pedicellatum were 91.20%, 9.45%, 1.52%, 8.58%, 63.97%, 36.88%, 10.81% and 37.52% DM, respectively. The ash content of Pennisetum pedicellatum (9.45% DM) recorded in the present study is in line with the observation of [51], who reported a value of 9.0%. The EE content of Pennisetum pedicellatum (1.52%) recorded in this study indicated that it was a low source of energy. It has been reported that the consumption of feeds and forages with low EE contents can increase methane production, which is detrimental to the environment and further increases energy inefficiency in ruminants [52]. In contrast, excessive consumption of EE in ruminant livestock may impair microbial activities, limiting fiber digestibility [49]. The CP values for Pennisetum pedicellatum (8.58% DM) observed in this study are not in agreement with the findings of [53], who reported 11% CP. Generally, the CP contents of the two cultivated forages in the present study are greater than the minimum level of 7–8% DM required for optimum rumen function and feed intake in ruminants [15]. However, it was lower than the recommended minimum requirements for lactation (12%) and growth (11.3% DM in ruminants) [16]

The NDF content of Pennisetum pedicellatum (63.97%) recorded in the present study is lower than the range of 72.78–77.68% reported by [54]. However, it is within the range of 58.82–63% [55]. The ADF values for Pennisetum pedicellatum (36.88%) recorded in this study are consistent with the results of [56], who reported values ranging from 16.63 to 36.14% DM. However, this value is greater than the reported range of 17–21% recommended for rumen stability [31]. The ADL content of cultivated forages ranged from 10.8% in Pennisetum pedicellatum at ML to 12.22% DM in Pennisetum purpureum at HL. The high contents of ADL in cultivated forages reported in this study could have a negative influence on digestibility, which causes a decrease in the availability of nutrients. Generally, the variation in chemical composition of cultivated forages recorded in the present study and in the literature might be due to differences in agro-ecological zone, plant species, soil on which they were grown, weather conditions during growth, and the stage of maturity.

3.4 Chemical composition of crop residues

Table 4 presents the mean (± SE) chemical composition of the crop residues in the three agro-ecological zones of the study area. The results revealed that, except for NDF, all chemical composition variables of maize stover were influenced by the AEZ (P < 0.05). The mean DM, Ash, EE, ADF, ADL and CF contents of maize stover were significantly (P < 0.05) greater in the HL than in the LL and ML AEZs. The mean DM, Ash, EE, CP and CF contents of teff straw differed significantly (P < 0.05) across the AEZs, with the highest values recorded in the HL AEZs compared to those in the ML and LL AEZs. The mean EE, CP and ADF contents of wheat straw were significantly (P < 0.05) greater in the HL than in the ML and LL AEZs. The mean DM and EE contents of barley straw were significantly (P < 0.05) greater in the HL AEZs than in the ML and LL AEZs. However, the NDF and ADF contents were significantly (P < 0.05) greater in the ML AEZs than in the LL and ML AEZs. There was no significant effect (P > 0.05) of AEZs on the DM, Ash, CP, ADF, ADL and CF contents of sorghum stover except for EE and NDF (P < 0.05), which were greater in the HL AEZs than in the ML and LL AEZs.

Table 4 Chemical composition (% DM) of crop residue from three AEZs in the Gera district, southwest Ethiopia
Full size table

The DM contents of maize stover, teff straw and barley straw (Hordeum vulgare L.) differed significantly (P < 0.05) across the AEZs. The DM content of the crop residues in the present study ranged from 89.47 ± 0.53 in teff (Eragrostis tef (Zucc) Trotter.) straw at LL to 94.28 ± 0.20 in maize (Zea mays L.) stover at the HL AEZ. The results of the present study are in agreement with the findings of previous studies [57, 58] reporting that the DM content of various crop residues ranged from 89.86 to 94.77% elsewhere in Ethiopia.

In this study, the ash content of maize stover and teff straw varied significantly (P < 0.05) in response to the AEZ. The average ash content of crop residues ranged from 6.94 ± 0.24% in sorghum stover at LL to 11.61 ± 0.57% DM in wheat straw at the HL AEZ. The mean ash content in sorghum stover, maize stover and teff straw recorded in the current study was higher than the ash values reported by [59].

There was a significant difference (P < 0.05) in the EE content of crop residues (P < 0.05) across the studied AEZs. The average EE content of crop residues ranged from 1.29 ± 0.08 in wheat straw at ML to 2.41 ± 0.09% in barley straw at HL AEZ. The EE contents of maize stover and wheat straw were greater in the HL treatment than in the ML treatment. However, the EE content of wheat (Triticum aestivum L.) straw was lower at the ML, and that of sorghum stover was lower at the LL AEZ.

The CP contents of maize stover, teff straw and wheat straw varied significantly (P < 0.05) across the AEZs. The CP content of crop residues ranged from 2.43 ± 0.13 for wheat straw at the ML to 4.81 ± 0.35 for barley straw at the HL AEZ. These low CP contents of crop residues might be attributed to the age of the crops that were harvested after the seeds had dried well before they were harvested. The CP content of crop residues recorded in the present study was much lower than the minimum level of 7% required for rumen microbial function [19]. The CP content of crop residues in this study was in agreement with the CP values of < 7% for maize stover, teff straw, wheat straw, barley straw and sorghum stover reported by [60]. Nasrullah [61] stated that voluntary feed intake decreases rapidly if the CP content of roughages is below 6.2% DM. Cereal straws generally have a low nitrogen content and are composed of cell wall components with little soluble cell content [62].

The NDF content of the crop residues did not vary significantly (P > 0.05) across the AEZs except for those of barley straw and sorghum stover. The NDF content in crop residues ranged between 70.61 ± 1.17 in sorghum stover at LL and 79.95 ± 3.02 in maize stover at HL. All crop residues reported in this study had NDF contents greater than 65% DM and were classified as low-quality roughages [35]. Similar to the findings of the present study, Sisay [63] also reported NDF contents of crop residues higher than 70% DM. It has been reported that as plants mature, the NDF, ADF and lignin contents increase, while the CP content decreases [64], emphasizing that an increase in these parameters is influenced by the maturity stage of the crop residues. Feds with NDF contents less than 45%, 45–65% and greater than 65% are considered to be of high, medium and low quality, respectively [65, 66]. A high NDF above 72% will cause a low intake of forage [67], and as NDF values increase, DM intake generally decreases [68].

The ADF contents of maize stover, wheat straw and barley straw varied significantly (P < 0.05) across the studied AEZs. The ADF content of crop residues ranged from 48.26 ± 0.38 in wheat straw at HL to 60.31 ± 1.61 in sorghum stover at the LL AEZ. A high ADF content in crop residues could result in lower digestibility since the digestibility of feed and its ADF content are negatively correlated [69]. Feeds with ADF contents less than 30% and greater than 40% are considered to be of high quality and poor quality, respectively [66]. The ADF content of crop residues in this study was greater than 48% DM, indicating their low digestibility.

Agro-ecology influenced the ADL content of maize stover only (P < 0.05). The ADL content of crop residues varied from 7.52 ± 0.49 in teff straw at HL to 13.34 ± 0.55 in wheat straw at ML. The findings of the present study are consistent with the results of [70], who reported ADL contents of 8.07, 9.22 and 10.30% DM for barley straw, teff straw and wheat straw, respectively. ADL represents an indigestible portion of rough materials and forms complexes with cellulose and hemicellulose constituents through lignification, thereby impairing microbial digestion. According to Van Soest [19], an ADL content above 6% has a negative impact on the digestibility of forage. Therefore, all crop residues included in this study had ADL contents above this recommended level, resulting in low digestibility by ruminants. Generally, the variation in the chemical composition of crop residues between the present study and the literature might be due to differences in the agro-ecological zone, climatic conditions, soil fertility on which the crop was grown, weather conditions during growth, and maturity stage.

4 Conclusions

The results revealed that variation in agro-ecological zone had effect on most of the evaluated chemical composition traits of the feedstuffs. This might be due to variations in microclimate, soil type and organic matter content, temperature, and rainfall, which might affect plant growth and maturation. The indigenous fodder tree and shrub species in both AEZs had greater CP and lower NDF, implying their potential to supplement poor quality feeds such as crop residues, stubble and dry natural pasture grasses, especially during the dry season. On the other hand, crop residues had the lowest CP below the minimum recommended level (7–8%) required for the activity of rumen microorganisms and maintenance requirements of ruminants and the highest NDF, implying the need to either improve their nutritive value through urea treatment or supplementing animals with protein source feeds. Generally, AEZ variation in the current study had an influence on the chemical composition traits of feed resources attributed to its effects on the microclimate conditions that affect plant growth and maturation in the study sites. It is recommended that evaluation of the in-vitro organic matter digestibility and mineral contents of these feed resources is essential to fully understand the nutritive values of these feed resources. In addition, animal-based feeding trials are needed to substantiate the findings of the current study on animal response.