Determination of feed balance for smallholder mixed crop–livestock farming system in Adwa district of Tigray, Northern Ethiopia

Abstract

A feed inventory and feed balance assessment study was conducted in Adwa district of central Tigray, Northern Ethiopia, with the objective to investigate available feed resources and evaluate livestock feed requirements under the smallholder mixed crop–livestock production system. Sixty livestock households were selected systematically from two farmer associations that were selected for their livestock potential. Livestock holding size, feed types, feed availability and feed demand were estimated using household interviews. Representative feed sampled were collected at field and analysed for dry matter (DM), crude protein (CP) and in vitro DM digestibility. Metabolizable energy (ME) and digestible CP (DCP) were estimated from equations. Feed balance was evaluated as the difference between feed supply and feed demand. The collected data were subjected to a descriptive statistical analysis. The major livestock feed resources available in the area included crop residues, crop stubble, grazing lands and mixed hay. On average, about 1.84 tonne DM feed, 13,635 MJ ME and 99.13 kg DCP were produced at farm level in year-round base. Crop residues (51%) was the dominant contributor of the annual feed resource base followed by grazing lands (34%), stubble grazing (11%) and mixed hay (4%) in descending order. A negative feed balance was seen for all nutrient types. The livestock are faced with critical deficiency of 56% DM, 52% ME and 57% DCP. This implies that the livestock production is practiced under annual feed deficit conditions.

Keywords:

• feed demand
• feed gap
• feed resources
• feed supply
• livestock

PUBLIC INTEREST STATEMENT

Livestock production is a key economic activity for the smallholder farmers in Adwa district of Northern Ethiopia. Feed situation is the main concern of the smallholder livestock production system. Understanding the feed type and feeding status of livestock is necessary to address feed shortage problem. Accurate scientific investigation of feed availability and livestock feed gap is considered as a prerequisite for making the best use of available resources. Livestock feed balance analysis is undertaken to determine the feed status at national and farm levels. This can suggest a clear strategy for the development of feed resources and their efficient utilization that guides the proper way of feed production, management and utilization.

1. Introduction

Ethiopia is an agrarian country owning huge livestock resources of different types. The country holds about 60.39 million cattle, 31.30 million sheep, 32.74 million goats and 56.06 million poultry (CSA, 2018). The livestock provide a large variety of goods and services like meat, milk, drought power and income. A large number of livestock is owned by smallholder farmers who practice animal rearing in fragment lands. Despite the fact, the livestock production is faced with many complicated challenges related with feed availability, disease occurrence and genetic potential. Feed shortage is the key problem in the operation of livestock production activities in Ethiopia (Shapiro et al., 2017). The feed deficit problem is more pronounced in the smallholder livestock production system. FAO (2018) reported a negative livestock feed balance in Ethiopia with deficit of 9% DM, 45% energy and 42% protein. Though many efforts have been made to address the livestock feed shortage problem, it has remained unsolved so far for various reasons.

Mixed crop–livestock production system is the dominant one in Tigray region including Adwa district which is the focus of this study. The mixed farming system sustains the life of many smallholder farmers in the region. There is a high inter-dependency between the two components of the system in which livestock element is interwoven with crop cultivation. Livestock are based on different feed resources such as crop residues, natural pastures, hay and non conventional feeds. The distribution and availability of these feed types vary with season and location (CSA, 2018). The contribution of the feeds is affected by farming system, agro-ecological zone and other specific factors. Livestock are more dependent on crop residues as feed resources in Tigray region (BoANRD, 1997; Yayneshet, 2010). The contribution of natural pastures to the annual feed supply is diminishing across time mainly due to land use change (Alemayehu, 2002). Crop residues are noted for their poor feeding value in terms of protein contents, energy value and digestibility (Adugna & Said, 1994; Yayneshet, 2010). Animals based on roughage feeds exclusively and/or with limited feeding level are not expected to achieve improved performance. A strategic supplementation and sufficient feeding regime is required to exploit the full genetic potential of the animals. This entails matching the supply and demand of dry matter (DM) and nutrients to explore the production and reproduction potential of the animals through achieving optimum feed balance.

A detailed study of livestock feed balance is required to identify feed types, quantify feed amount and estimate feed requirement of animals in order to evaluate the feed gap or surplus (FAO, 2012). A negative feed balance can disrupt the normal function of the mixed crop–livestock farming system, causing low crop production, reduced livestock performance, and can adversely affect the natural resources management. Checking the livestock feed balance is mandatory to monitor the existing functioning of the system and to adjust to the dynamic socioeconomic conditions. Many scientific evidences noted the importance of feed balance analysis for formulating and planning feed improvement policy and strategy (FAO, 2018; Mekete et al., 2018; Solomon, 2004; Zewdie et al., 2011). This enables to analyse the feed and feeding conditions under which the livestock production is operated by livestock producers. This does mean that feed balance status can serve as potential indicator to examine sustainability of farms. This makes it vital to establish baseline data on the types and quantity of feed available at farm level. According to FAO (2018), information on feed inventory and feed balance would be helpful to formulate and implement sustainable livestock development programs. With this insight, the current study was carried out with the aim to investigate the major feed resources available and evaluate the livestock feed balance under the mixed crop–livestock farming system in Adwa district of central Tigray, Northern Ethiopia.

2. Materials and methods

2.1. Study area

The research work was conducted in Adwa district of central zone of Tigray, Northern Ethiopia. The study area is situated at 14° 10’ N latitude and 38° 54’ E longitude, with average elevation of 1,907 meters above sea level. The area receives an annual precipitation amount of 600 to 850 mm. The mean minimum and maximum temperatures are 12°C and 27°C, respectively. Mixed crop–livestock production system is the dominant farming form practiced by the resource poor smallholder farmers. Crops like Teff (Eragrostis tef), wheat, barley, Hanfets, maize and pulses are dominantly produced by the smallholder farmers. Hanfets is a cereal crop mixture of wheat and barley grown together. Livestock are integral components of the farming system offering many benefits to the smallholder farmers. The smallholders raise different livestock types including cattle, goats, sheep, donkeys and chickens (WAOARD, 2016).

2.2. Sampling and data collection methods

2.2.1. Household interviews

The study was carried out in Adwa district from February to November, 2013. A multistage sampling technique was employed to select sample areas and households. First, two farmer associations (Debre-Genet and Laki’a) were purposely selected based on their potential representativeness of the mixed crop–livestock farming system. Second, 60 livestock households (30 from each farmer association) were chosen using simple random sampling methods. The participating farmers were selected from the list of population frame obtained from peasant association offices. Data were gathered from the selected households using household interviews and participatory rural appraisal tools. A semi-structured questionnaire was prepared, tested and refined to collect data using face-to-face interviews. The pretested questionnaire included data on household characteristics, land holding size, livestock ownership, livestock species composition, livestock herd structures, feed inventory, feed production, private grazing land size, types of crop produced, area cultivated for each crop type, crop yield and other relevant data at farm level. Each household was taken as a unit of analysis. Focus group discussions were held with farmers at each of the selected peasant associations. Secondary data and information were also extracted from different published and unpublished sources. Total household livestock holding size that obtained from household interviews were converted to tropical livestock unit (TLU, weight of 250 kg) with multiplier conversion factor of 0.7 for cattle, 0.1 sheep, 0.1 goat, 0.5 donkey and 0.01 chicken as described by Jahnke (1982) and Gryseels (1988).

2.2.2. Estimation of annual feed production

The annual feed resources required for animals came from different sources. It is necessary to quantify the feed supplied by each to estimate its proportional contribution to the annual feed pool. The amount of feed obtained from each source was quantified using proper techniques and procedures. The feed amount in DM that can be contributed by the crop residues was estimated from grain yield using a grain-to-residues conversion factor (FAO, 1987; Kossila, 1988) in which multipliers of 1.5 for Teff, wheat, barley and Hanfets; 2 for sorghum and maize; and 1.2 for pulse and oil crops were used. It was assumed that about 90% of the residues is efficiently utilized as animal feed, while the other 10% is wasted during collection and feed process and/or used for other purposes (Adugna & Said, 1994). DM that potentially can be obtained from stubble grazing was estimated by multiplying the average household cultivated land holding with 0.5 tonne DM per ha based on FAO (1987). Similarly, DM yield from private grazing land was estimated by multiplying the average household private grazing land holding with the annual DM output of natural pastures which is taken as 2 tonnes DM per hectare based on previous studies and documented information (FAO, 1987; MoA, 1989). The private grazing land size was obtained from the household interview, while communal grazing was shared from accessible communal grazing lands. A livestock density was used to estimate the share of communal grazing per individual household. As a result, a livestock density of 15.57 TLU/ha was derived from livestock number and accessible grazing size. Thereafter, the livestock density was used to allocate the communal grazing to each livestock holder. Utilization factor of 75% was used to determine the available feed resources from extensive grazing (WBISPP, 2002). Available feed DM from hay was estimated using a formula recommended by Fourth Livestock Development Project (MoA, 1989) and household interviews. Feeds that can be potentially available from browse plants, household wastes, Atella (residues of local brewery), mill by-products, kitchen wastes and the like were not considered in the feed supply estimation.

2.2.3. Estimation of annual nutrient supply

The supply of available energy was measured in metabolizable energy (ME) which was calculated from in vitro DM digestibility (IVDMD). Similarly, the supply of protein nutrient was measured in the form of digestible crude protein (DCP) that was computed from crude protein (CP). The DM and nutrients supplied by each feed type were calculated as total DM output multiplied with nutrient concentration of respective feed (Abdinasir, 2000).

2.2.4. Estimation of annual DM and nutrient requirements

DM and nutrient (ME and DCP) requirements were determined following the recommended scientific procedures. One TLU animal consumes about 2.5% of its 250 kg live body weight on daily basis (Jahnke, 1982; Kearl, 1982), making the daily DM requirement of 6.25 kg DM per day. This makes the annual DM requirement to be about 2,280 kg. A ME of 118.00, 93.00 and 103.45 kcal per metabolic weight per day was used for the maintenance of cattle, sheep and goats, respectively. Similarly, a DCP of 2.86, 1.72 and 2.51 g per metabolic weight per day was used for the maintenance of cattle, sheep and goats, respectively. Also, 14.90 MJ ME and 0.18 kg DCP was used for the daily maintenance requirement of donkeys as recommended by McCarthy (1986) as cited in Zewdie et al. (2011). The annual nutrient (ME and DCP) requirement for maintenance was computed as the daily maintenance requirement of all animal types multiplied with 365 days (Kearl, 1982). The total household livestock TLU was obtained from the household interview.

2.3. Sample feed preparation and chemical analysis

Feed samples were collected from major feed resources such as residues of different crops, mixed hay and green grass. The field collection of sample feeds was performed using paper bag that is labeled carefully to have full information. DM and nitrogen were determined as per the procedures described by AOAC (1990). CP was calculated as N × 6.25. IVDMD was analyzed by the Tilley and Terry method as modified by Van Soest and Robertson (1985). ME was calculated from IVDMD using a formula: ME (MJ/kg DM) = 0.17*IVDMD (%) − 2.0 (AAC, 1990). Similarly, DCP was computed from CP using an equation: DCP (g) = 0.929*CP (g) − 3.48 (Church & Pond, 1982).

2.4. Feed balance analysis

The total feed supply was calculated as the summation of crop residues, grazing lands, stubble grazing and mixed hay. Similarly, the total feed demand was determined by summing up the feed requirement of each animal species: cattle, sheep, goats and donkeys. The livestock feed balance was calculated at individual household level as the difference between the annual feed DM, ME and DCP supply and the annual feed DM, ME and DCP demand.

2.5. Data statistical analysis

Data were subject to the descriptive statistics of the SPSS (2013, Version 22) computer statistical software. Descriptive statistics such as frequency, percentage, mean and standard error were used to describe variables and summarize the results.

3. Results

3.1. Household socioeconomic characteristics

Socioeconomic characteristics of the sampled households are presented in Table 1. The study included 85% male households and 15% female households. The average age of households was 42 years with a range of 21–70 years. Majority of households (60%) fall in the age category of 31–50 years, while others (37%) in 51–70 years. The study found an average family size of 4.6 persons per household. Majority of households (62%) owned 5–8 family members, while 15% of them had 1–4 persons and 23% with 9–12 persons. Most of the livestock farmers were educated while few were illiterate.

Table 1. Socioeconomic characteristics of sampled households in Adwa district of central Tigray, Northern Ethiopia

Socioeconomic variables	Number (n)	Percent (%)
Household sex
Male	51	85
Female	9	15
Household age
20–30 years	2	3
31–50 years	36	60
51–70 years	22	37
Household family size
1–4 persons	9	15
5–8 persons	37	62
9–12 persons	14	23
Household education level
Illiterate	16	27
Primary school (1–6 grade)	34	56
Secondary school (7–10 grade)	10	17

3.2. Landholding and its use

The landholding and land use pattern is shown in Table 2. The average landholding size was 0.59 ± 0.37 ha per household. The livestock households allocated the land predominantly for the purpose of crop cultivation (0.53 ± 0.02 ha), accounting for about 90% of the total landholding. The farming land was used to grow different crop types like Teff (Eragrostis tef), wheat, barley, Hanfets, maize, finger millet and pulses. Private grazing land (0.04 ± 0.01 ha) and irrigated land (0.02 ± 0.01 ha) shared less land with 7% and 3%, respectively. Majority of the sampled households (67%) owned 0.1–0.5 ha, 26% with 0.6–1.0 ha, 7% with 1.1–1.5 ha and no farmer was observed having above 1.5 ha land. Similarly, no farmer was observed growing improved fodder.

Table 2. Land holding size and land use pattern per household in Adwa district of Tigray region, Northern Ethiopia

Land size (ha/HH)	Frequency	Percent	Land type	Mean ± SE (ha)	Land share (%)
0.01–0.50	40	67	Crop land	0.53±0.02	84
0.51–1.00	16	26	Grazing land	0.04±0.00	7
1.01–1.50	4	7	Irrigated land	0.02±0.00	3
Total	60	100	Total land	0.59±0.37	100

HH = household; ha = hectare; SE = standard error.

3.3. Livestock ownership and herd composition

Livestock are important assets for the smallholder farmers of the mixed system. The average livestock holding per household is depicted in Table 3 in heads and TLUs. The average household livestock holding was found to be 3.21 ± 0.17 TLU. The livestock herd was dominated by cattle (2.33 ± 0.12 TLU) followed by donkeys (0.36 ± 0.05 TLU), sheep (0.28 ± 0.04 TLU) and goats (0.24 ± 0.04 TLU) in descending order. This can be evidenced by the high proportional contribution of cattle at 73%. The percentage share of donkeys, sheep and goats in the livestock herd amounted at 11%, 9% and 7%, respectively. The cattle herd was composed of indigenous breeds (97%), while exotic/crossbred accounted for only 3%.

Table 3. Livestock holding per household in Adwa district, Northern Ethiopia (mean±se)

Animal type	Head	TLU	TLU share (%)
Cattle	3.33±0.03	2.33±0.12	73
Sheep	2.78±0.45	0.28±0.04	9
Goat	2.37±0.41	0.24±0.04	7
Donkey	0.72±0.10	0.36±0.05	11
Total	-	3.21±0.17	100

TLU = tropical livestock unit (weight of 250 kg); SE = standard error.

3.4. Availability of major feed resources

Estimated annual feed supply from major feed resources at household level is presented in Table 4. On average, about 1.84 tonnes DM feed was produced at farm level in year round base. The feeds were majorly contributed from crop residues (0.93 tonnes DM) followed by grazing land (0.63 tonnes DM), stubble grazing (0.20 tonnes DM) and hay (0.08 tonnes DM) in descending order. The proportional contribution of these feeds is 51%, 34%, 11% and 4%, respectively. The available crop residues were derived from residues of Teff, wheat, Hanfets, maize, barley, finger millet, pulses and sorghum in the order of their importance. The contribution of browse plants and other non-conventional feeds like Atella (residue of local liquor), household wastes and mill by-products was not implicitly considered, and this may underestimate the quantified feed supply. It is difficult to quantify them at household level as the respondents have difficulty in recalling back such feeds given to animals.

Table 4. Annual feed supply from major feed resources at household level in Adwa district of central Tigray, Northern Ethiopia

Feed type	Feed supply (tonne DM/year)	Contribution (%)
Crop residues	0.933	51.00
Teff straw	0.342	18.61
Wheat straw	0.182	9.90
Hanfets straw	0.139	7.56
Maize stover	0.083	4.52
Barley straw	0.073	3.97
Finger millet straw	0.057	3.10
Pulse straw	0.038	2.07
Sorghum stover	0.013	0.71
Oil crop straw	0.004	0.22
Grazing land	0.630	34.28
Stubble grazing	0.199	10.83
Natural mixed hay	0.078	4.24
Total	1.838	100.00

DM = dry matter.

3.5. Nutritional quality of major feed resources

The nutritional composition of major feed resources in terms of IVDMD, ME, CP and DCP is presented in Table 5. As it can be seen from the table, oil seed straw (9.28 MJ/kg DM) has higher ME followed by Teff straw (8.56 MJ/kg DM), pulse straw (8.39 MJ/kg DM), hay (7.82 MJ/kg DM) and green grass (7.42 MJ/kg DM). Sorghum stover (7.68 MJ/kg DM), barley straw (6.87 MJ/kg DM), Hanfets straw (6.57 MJ/kg DM), wheat straw (6.46 MJ/kg DM) and maize stover (6.42 MJ/kg DM) contain ME value in descending order. DCP was greater for green grass (141.82 g/kg DM) followed by oil seed straw (81.80 g/kg DM), pulse straw (71.86 g/kg DM) and hay (66.67 g/kg DM) in reducing order. Barley straw (46.78 g/kg DM), Hanfets straw (26.06 g/kg DM), maize stover (25.97 g/kg DM), Teff straw (25.32 g/kg DM), wheat straw (22.16 g/kg DM) and sorghum stover (22.16 g/kg DM) have lower DCP in that order.

Table 5. Energy and protein contents of major feed resources in Adwa district of central Tigray, Northern Ethiopia

Feed type	IVDMD (% of DM)	ME (MJ/kg DM)	CP (% of DM)	DCP (g/kg DM)
Teff straw	62.10	8.56	3.10	25.32
Wheat straw	49.77	6.46	2.76	22.16
Hanfets straw	50.44	6.57	3.18	26.06
Maize stover	49.53	6.42	3.17	25.97
Barley straw	52.15	6.87	5.41	46.78
Pulse straw	61.09	8.39	8.11	71.86
Sorghum stover	56.97	7.68	2.76	22.16
Oil seed straw	66.33	9.28	9.18	81.80
Hay	57.74	7.82	7.55	66.67
Green grass	55.40	7.42	15.64	141.82

DM = dry matter; CP = crude protein; IVDMD = in vitro dry matter digestibility; ME = metabolizable energy; MJ = mega joule; DCP = digestible crude protein.

3.6. Nutrient supply from major feed resources

Livestock feeds are sourced from crop residues, grazing lands, stubble grazing and hay. Annual farm-level produced nutrients are estimated in terms of ME and DCP (Table 6). The annual ME produced at farm level was estimated at 13,635 MJ. The energy was largely derived from crop residues (6,874 MJ) followed by grazing lands (4,675 MJ), stubble grazing (1,477 MJ) and mixed hay (610 MJ) in that order. The proportional contribution of these listed feeds in terms of energy was 50%, 34%, 11% and 5%, respectively. The energy available from crop residues significantly comes from Teff straw (21%) followed by wheat straw (8%), Hanfets straw (7%), maize stover (4%), barley straw (4%), finger millet straw (3%), pulse straw (2%) and sorghum stover (1%) in the order of their importance. Farm-level estimated DCP was 99.13 kg, which was significantly derived from grazing lands (50.55 kg) followed by crop residues (27.41 kg), stubble grazing (15.92 kg) and mixed hay (5.20 kg) in descending order. The listed feeds contributed about 51%, 28%, 16% and 5% of the total protein outputs, respectively.

Table 6. Annual nutrient supply from major feed resources at farm level in Adwa district, Northern Ethiopia

Feed type	ME (MJ)	ME share (%)	DCP (kg)	DCP share (%)
Crop residues	6,873.70	50.00	27.41	27.65
Teff straw	2,927.52	21.47	8.66	8.74
Wheat straw	1,175.72	8.62	4.03	4.07
Hanfets straw	913.23	6.70	3.62	3.65
Barley straw	532.86	3.91	2.16	2.17
Maize stover	501.51	3.68	3.41	3.44
Finger millet straw	367.08	2.69	2.18	2.20
Pulse straw	318.82	2.34	2.73	2.75
Sorghum stover	99.84	0.73	0.29	0.29
Oil seed straw	37.12	0.27	0.33	0.33
Grazing land	4,674.60	34.28	50.55	50.99
Stubble grazing	1,476.58	10.83	15.92	16.11
Natural mixed hay	609.96	4.47	5.20	5.25
Total supply	13,634.84	100.00	99.13	100.00

ME = metabolizable energy; MJ = mega joule; DCP = digestible crude protein.

3.7. Estimation of DM and nutrient requirement

Estimated annual nutrient requirement for maintenance at household level is presented in Table 7 in terms of DM, ME and DCP. As shown in the table, the annual DM, ME and DCP requirements were estimated at 4.21 tonnes, 28,486 MJ and 231 kg, respectively. The energy requirement was higher for cattle (90%) followed by donkey with 8% and small ruminants with 2%. Similarly, protein requirement was more for cattle (90%) followed by donkeys (9%) and small ruminants (1%).

Table 7. Annual maintenance requirement of livestock types per household in Adwa district, Northern Ethiopia

Animal type	DM (t)	Share (%)	ME (MJ)	Share (%)	DCP (kg)	Share (%)
Cattle	3.75	89.07	25,650.74	90.05	209.35	90.46
Sheep	0.07	1.66	439.46	1.54	1.56	0.67
Goat	0.06	1.43	438.00	1.54	0.51	0.22
Donkey	0.33	7.84	1,957.86	6.87	20.01	8.65
Total requirement	4.21	100.00	28,486.06	100.00	231.42	100.00

DM = dry matter; ME = metabolizable energy; MJ = mega joule; DCP = digestible crude protein.

3.8. Livestock feed balance analysis

Livestock feed balance is calculated by comparing feed requirement with the existing feed supply. The feed balance is presented in Table 8 in the form of DM, ME and DCP. As shown in the table, farm-level annual DM requirement was 4.21 tonnes, while the annual DM production is 1.84 tonnes, showing a deficit of 2.37 tonnes DM per year (56%). With the same calculation, the annual energy deficit was 14,851 MJ ME (52%) considering the annual demand (28,486 MJ ME) and annual supply (13,635 MJ ME). Similarly, the mismatch between the annual protein requirement (231 kg DCP) and the annual protein supply (99 kg DCP) resulted in a shortage of 132 kg DCP (57%). Protein deficiency was slightly higher as compared to the deficiency of energy.

Table 8. Livestock feed balance at farm level in year base in Adwa district of central Tigray, Northern Ethiopia

Major nutrients	Feed demand	Feed supply	Feed balance	Feed deficit (%)
DM (t)	4.21	1.84	−2.37	56.29
ME (MJ)	28486.06	13634.84	−14851.22	52.14
DCP (kg)	231.42	99.13	−132.29	57.16

DM = dry matter; ME = metabolizable energy; DCP = digestible crude protein.

4. Discussion

4.1. Land ownership and livestock holding

The average household land holding size (0.59 ha) in the current study is comparative with the survey report of CSA (2001) at 0.56 ha, but it is lower than the regional (1.15 ha) and national (1.48 ha) average values reported by Central Statistical Authority (CSA) and World Bank (2017). A higher land holding (1.44 ha/HH) was reported in Tanqua-Abergelle district of central Tigray, Northern Ethiopia (Tikabo & Shumuye, 2016). Land is largely devoted for cereal crop production. No farmer was observed practicing fallowing and forage cultivation, indicating the serious scarcity of farm land in the area. The limited private grazing land availability also reflects the critical land shortage. Similarly, the present livestock holding size (3.21 TLU) was lower as compared to the findings of Tikabo and Shumuye (2016), 4.93 TLU, and Aledin et al. (2019) who found 7.22 TLU in north western zone of Tigray. More livestock holding was reported by Solomon (2004), 10.09 TLU/HH, in Bale highlands of central Ethiopia. The apparent variation in livestock holding could be related with land size and feed availability. The livestock herd is largely composed of cattle, indicating the paramount importance of cattle in the mixed farming system. The cattle rearing is operated using local breeds, known as Arado cattle, which is essentially raised for drought power, particularly land tillage. In favor of this study, CSA (2018) reported that nearly 98% Ethiopian cattle population are indigenous breeds, while exotic/hybrid accounts for only 2%. All the sheep and goats are indigenous type and no farmer was observed rearing exotic or crossbred one.

4.2. Estimation of feed and nutrient supply

In the present study, crop residues, grazing lands, stubble grazing and natural mixed hay are investigated as the major feed resource bases contributing to the annual feed DM and nutrient supply. Crop residues contribute more to the annual feed supply in terms of DM, energy and protein. This implies that crop residues are important feed resources in the area where mixed crop–livestock farming system is dominantly practiced by smallholders. A large proportion of land is utilized for crop growing than other land use type including grazing areas. The available size of private grazing land holding is very limited, indicating the serious scarcity of land in the area. In addition to this, fallow lands as source of livestock feeds were non-existent, and this shows the scarcity of crop lands. This study agrees with a survey conducted in central highlands of Ethiopia where crop residues contributed 50–80% of the total feed supply for animals (Ahmed et al., 2010). This as it is, the proportional contribution of crop residues in the present study (51%) is higher than the regional average of 41% recently reported by CSA (2018) for Tigray region. The present crop residues production (0.933 tonnes DM/HH) is lower than that of Solomon et al. (2008) who reported 6.70 tonnes DM/HH in Bale highlands of Ethiopia. The variation in household level crop residues production could be due to difference in land size, crop type, cropping pattern, rainfall amount, soil type, input use and other factors.

The contribution of grazing areas to the feed pool is low (34%) as compared to the regional average of 40% reported by CSA (2018). The contribution of the feeds varies with time as there is dynamism in the farming system related with demographic change, settlement and urbanization. The limited contribution of grazing lands can be explained by change in land use pattern where grazing areas are converted into cropping lands to produce more grain food for the ever-increasing human population, particularly in highland areas (Alemayehu, 2002). In favor of this fact, Solomon et al. (2008) noted the progressive expansions of crop lands at the expense of grazing lands, resulting in increased use of crop residues as the major feed source in the mixed farming system of Bale highlands of central Ethiopia. This implies that the fate of grazing lands is not certain and thus, it is impractical to improve feed availability through expansion of grazing lands (Zewdie et al., 2011). This as it is, natural pastures are still the dominant contributor of the livestock feed resource base at national level with an average value of 56% followed by crop residues (30%) (CSA, 2018). The feed supply estimation did not consider the feeds that can be made available from non-convention feed resources such as household wastes, Atella, mill by-products and kitchen wastes for the fact that the occasional availability of these feeds makes it difficult to quantify these feeds at farm level. This might have led to an underestimation of the annual feed supply at farm level. An area is characterized as mixed crop–livestock system if more than 10% of the animal demand is derived from crop residues (Sere & Steinfeld, 1996). Based on this fact, the study area is classified as mixed farming system with high interweaving between crop component and livestock element.

In terms of nutrient supply, crop residues are the major contributors of energy (50%). This was more or less similar with the findings of Abdinasir (2000) and Solomon (2004) in which the share of crop residues was reported to be 45–55% of the total ME available at household level in the central highlands of Ethiopia. This shows that crop residues are the dominant sources for the available nutrients. The contribution of other feed sources like grazing areas is low compared to residues, which is different from other studies. Unlike this study, natural pastures were appreciated as the major source of feeds for ruminant animals in other parts of Ethiopia (Abdinasir, 2000; Solomon, 2004). The fluctuation in proportional contribution of specific feeds to the annual feed pool with location can be attributed to the variation in land holding size, land use pattern, crop type and variety, cropping pattern, quality of feeds and other factors.

4.3. Livestock feed balance analysis

The current study revealed that there is a negative feed balance in the study district in terms of DM, energy and protein. This implies that the livestock production is operated under a feed shortage condition. Many research works concluded a negative livestock feed balance under the smallholder farmers of Ethiopia. The feed deficit evaluated in this study is more pronounced as compared to previous studies (CSA, 2013; Mekete et al., 2018; Zewdie et al., 2011). The variation in the extent of feed gap could be attributed to land size, livestock holding, feed availability, climatic conditions, land use pattern and other factors. Many scientific studies reported a negative feed balance under the smallholder livestock production system (FAO, 2018; Mekete et al., 2018; Zewdie et al., 2011). The recent report of FAO (2018) found feed deficit of 17% DM, 51% ME and 49% CP in Tigray region of Ethiopia. Similarly, about 42% feed DM deficit was reported by CSA (2013) at national level in Ethiopia. Zewdie et al. (2011) found that the annual feed supply only satisfies 64% DM, 81% ME and 66% DCP as maintenance requirements of the animals per farm in the central rift valley of Ethiopia. Mekete et al. (2018) also reported 51% DM, 19% ME and 38% DCP deficiency at farm level in the central highlands of Ethiopia. The feed gap obtained in the present study might have been overestimated as the contribution of the non-conventional feeds was not considered in the estimation of feed supply. Protein deficiency is higher when compared with energy deficiency. This implies that the livestock suffer from critical deficiency of protein, which would have a significant impact on the performance and health of livestock.

The negative feed balance in the present study indicates a mismatch between feed supply and feed demand in the mixed farming system. This can be evidenced by the prevailing low livestock production and productivity. It is obvious that the feed deficit will have a negative impact on the survival, growth and performance of animals. It becomes difficult to exploit the real production potential of the animals under the existing feeding conditions. This calls for narrowing the feed gap through matching the demand and supply of feeds to improve livestock productivity in a sustainable way. Strategic intervention options that can improve feed availability and quality need to be devised considering the prevailing conditions. Improving utilization and conservation of crop residues, using alternative feed resources, introducing adaptive improved forage species, introducing multipurpose legume trees, strategic supplementation of animals during dry seasons and other mechanisms need to be considered as feed development improvement options to reduce the impact of inadequate feeding on livestock performance and health status. In addition to this, improvement of crop residues in terms of quantity and quality should be considered in crop research programs. Beyond this, efforts must be exerted in improving the productivity, management and utilization of the existing grazing lands.

5. Conclusions

Crop residues, grazing lands, crop stubble and hay were identified as the major feed resources in the study area. The livestock feed balance was evaluated as negative status, indicating the prevalence of critical feed shortage in the system to satisfy the DM, energy and protein requirements of the current livestock holding at farm level. There is a mismatch between the annual feed supply and requirements of animals at present. This implies that the livestock production is being operated under critical annual feed deficit. It is a well-established fact that inadequate feed has negative impact on livestock performance and health conditions. It is unthinkable to explore the full genetic potential of the animal breeds with the available feeding situation. This entails strategic options to alleviate the feed gap through increasing feed availability in both quantity and quality.

Acknowledgments

We are very much pleased with Mekelle University for the financial support (CDANR/RB/6/2012) to conduct this study. We are also grateful to Abergelle Agricultural Research Centre for the logistic support. Our appreciation goes to the local farmers who made the study successful through providing the required information.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

The work was supported by the Mekelle University [CDANR/RB/6/2012].

Notes on contributors

Tikabo Gebremariam

Tikabo Gebremariam is working with an academic rank of Associate Professor at the Department of Animal, Rangelands and Wildlife Sciences, Mekelle University, Northern Ethiopia. He has been engaged in teaching and research activities for the last 20 years. His research interests focus on animal production, animal nutrition, forage production and meat sciences.

Shumuye Belay

Shumuye Belay is serving as a full time researcher at Mekelle Agricultural Research Center, Tigray Agricultural Research Institute, Northern Ethiopia. He has been working as an extension expert and researcher for the last 15 years. Currently, he is a PhD fellow at Addis Ababa University, Ethiopia. His key research interests are animal production, poultry production and animal genetics.