Livestock productivity, profitability and sustainability are limited by issues such as animal nutrition, genetics, disease, and management (1). However, a systematic way to estimate losses (such as mortality and loss in production) associated with these issues is lacking. Methods are needed to allow quantification of economic losses associated with livestock disease and other issues that depress livestock productivity, and to attribute these losses to specific causes or syndromes (2). Such information can then be used by policymakers to inform targeted investments in animal health.

Characterizing livestock production systems is key to understanding the burden of animal diseases. The burden of disease varies between livestock production systems due to differences in production purposes, breeds, management, and access to veterinary services. These factors all contribute to the risk of disease and the magnitude of losses (3).

Livestock populations are usually described in terms of the number of animals, yet this metric fails to account for differences in production purpose, breed, and population structure, which contribute to productivity and economic value. Using livestock biomass (i.e., the sum of individual liveweights for a given population) as an alternative allows a comparison of populations with different animal sizes across production systems. Furthermore, quantifying the current productivity (e.g., milk yields) of livestock in different production systems is the first step in understanding yield gaps. Similarly, estimating the economic value of livestock in a production system provides the basis for further work to capture expenditure on animal health. A recent study by Jemberu et al. (4) analyzed the biomass and economic value of small ruminants in production systems in Ethiopia. However, relevant information about cattle production systems in Ethiopia is still lacking.

Cattle make a critical contribution to livelihoods across Ethiopia, and the cattle population has increased from 53 million head in 2010 to 70 million in 2020 in response to increasing demand for livestock products (5). The vast majority (98%) of cattle in Ethiopia are local Bos indicus breeds kept by smallholder farmers for draft power and milk production in the mixed crop-livestock (70%) and pastoral (28%) zones (Shapiro 2017). The remaining cattle are crossbred or exotic breeds managed in specialized dairy and beef production systems in urban and peri-urban areas (6).

While the importance of cattle to Ethiopia is well documented, a lack of system-specific information on the biomass, productivity and value of cattle limits our ability to accurately analyze these systems. The aim of this paper is to address these knowledge gaps for the different cattle production systems within Ethiopia. This information will be used as a baseline for further disease burden analysis, which quantifies the scale and cost of production losses.

The biomass, productivity and value of cattle for 2021 were calculated for the lowland grazing (pastoral), mixed crop-livestock (CLM) and specialized systems using existing secondary data obtained from national statistics (7–9) and the literature. Analysis of the specialized systems was limited by the available data, and only specialized dairy systems could be considered.

Livestock biomass was calculated as the sum of the liveweight of all cattle within each production system using data on livestock populations, herd dynamics and liveweights. This information was combined with data on productivity and value to calculate livestock outputs and the value of livestock production. Cattle values include the stock value and the output value. In this study, the stock value refers to the total economic value of a cattle population in a system, and it was calculated by the average price of an individual multiply by the total population of the interested population. The output value refers to the total economic values of all the cattle products such as beef, milk and hides. The total value of each output can be calculated by using the total volume multiply by the average price of a product. The biomass and values were calculated by a year basis.

An understanding of the biomass and values of livestock by production system can benefit the development of good animal health policy. Knowledge of livestock biomass and the economic value of the livestock sector can help policymakers advocate for fair allocation of resources toward improving animal health and productivity. To the best knowledge of the authors, this is the first study describing the cattle biomass and economic value in the three dominant production systems in Ethiopia. Although cattle biomass in Ethiopia was calculated using estimates for cattle population and live weight in a previous study (29), the authors did not consider the variation in live weights of different age/sex groups of cattle, nor did they disaggregate their analysis by production system. The cattle population (25.8 million) has also increased substantially since that study was conducted. The findings from the present study offer insights on the difference of cattle production systems in terms of their population distribution, productivity, and values.

Ethiopian cattle production is constrained by low productivity. The indigenous cattle breed with small live body weights and poor milk productivity are dominant in the cattle population in this country. The vast majority (98%) of cattle in Ethiopia are local Bos indicus breeds kept by smallholder farmers for draft power and milk production in the mixed crop-livestock (70%) and pastoral (28%) zones (13). The remaining cattle are crossbred or exotic breeds managed in specialized dairy and beef production systems in urban and peri-urban areas. There is a great potential to improve the productivity in local cattle production systems. For example, the average live weight of a cow in the mixed and pastoral systems is only 242 kg in Ethiopia, while the average live weight of a cow in South Africa can reach to nearly 500 kg (30). Furthermore, in 2021, the annual milk output of an Ethiopian cow was substantially lower than the world average (449.9 kg VS. 2,692 kg) (31). These disparities raise significant concerns regarding the productivity of indigenous cattle breeds and the potential hindrance they pose in meeting the increasing demand for beef and dairy consumption within the country. Addressing the low offtake rates also holds relevance in evaluating the overall efficiency and growth potential of the cattle production systems in this country. It worth mentioning that the low productivity of local cattle was not only due to genetics but also other constrains, such as diseases, poor management and limited feed resources (1, 21, 32).

The output values per kilogram of biomass offer a valuable means of comparing different production systems. Notably, the dairy production system exhibits over twice the output values per kilogram of biomass compared to the other two production systems in Ethiopia, underscoring its higher productivity in this country. Moreover, this metric presents a benchmark for potential enhancements across all production systems, encouraging further efficiency and effectiveness in the livestock sector. Additionally, livestock biomass per capita serves as a significant measure of food security status and progress within a country. By assessing the per capita biomass, we can gain insights into the availability and utilization of livestock resources, thus influencing strategies for sustainable agricultural development. Furthermore, economic loss per kilogram of biomass emerges as an informative indicator for evaluating the health status of livestock populations. Analyzing economic losses associated with each unit of biomass sheds light on the overall well-being and would allow to compare disease burdens in different production systems and in different countries. The data presented in this study offer a baseline for disease burden analyzes by the GBADs program (2) and further analyzes by other stakeholders. For example, livestock biomass can also be used as an input into analysis of greenhouse gas emissions, land carrying capacity, and antimicrobial usage (33, 34). The GBADs program uses total biomass and economic value of livestock as the first step for disease burden analysis. Biomass is used as the denominator for comparing investments into livestock production, the value of livestock and their products, and the cost of lost production between different species and geographies. Biomass is more useful denominator than livestock population as it accounts for the variation in sizes of different livestock types (e.g., age, breed, species) and therefore also variations in inputs such as feed and antibiotics, and the outputs of products such as meat and milk. For example, Jemberu et al. (4) reported that the total biomass of the small ruminants was 2.1 billion kg in 2020/21 in Ethiopia. This study indicates that cattle biomass was seven times of biomass of small ruminants in the country.

The differences in population composition, reproductivity, productivity and economic values between the pastoral, CLM and specialized systems highlight the need to analyze disease burdens by production system. For example, cattle density was higher in the highland districts of the CLM production system than the lowland districts dominated by the pastoral (Figure 1). This is likely to impact the incidence of diseases spread by direct contact, droplets or airborne transmission (e.g., respiratory infections, foot and mouth disease), with Meadows et al. (35) reporting a positive association between diseases and cattle density. Similarly, differences in herd structure and production purpose will be associated with differences in the types of diseases and scale of economic impact. For example, farms in the CLM production system have the highest proportion of oxen, with draft power provided by these animals contributing almost a third of the total gross output value of livestock in the country (36). Health problems such as lameness that impact the utility of oxen would have a large impact on livelihoods of smallholder farmers who rely on these cattle for draft power. In addition, lameness in oxen in sowing and harvest seasons has potential implications for crop production and income of local farmers. In comparison, the specialized dairy farms have the highest proportion of milking cows, so they are more sensitive to diseases that would impact reproductivity and milk productivity, quality and safety (e.g., brucellosis, mastitis, salmonella).

Data on cattle populations and value reflect the different purposes of production in each system, and when combined with information on small ruminant populations (4) provide insights into the importance of different livestock across Ethiopia. The CLM system has the largest cattle population (i.e., 55 million head, 78% national cattle population in 2021) and cattle biomass (i.e., 11.3 million tonnes, 77% total national biomass in 2021), and the value of cattle production is concentrated in this region (i.e., 8.52 billion USD, 71% of total cattle output value in Ethiopia in 2021) (Table 4). Comparison with the results provided by Jemberu et al. (4) shows that cattle are also the most important livestock species in the CLM, with the annual value of small ruminants in 2021 being roughly one third that of cattle (3.25 billion USD). The dominance of small herd sizes in this system (e.g., cattle 4.4 heads, sheep 3 head, goats 4 head; Figure 2 and Jemberu et al. (4)) indicates the importance of ruminants to smallholder farmers as a source of draft power, income, and food. In the arid lowlands, cattle play a different but again central role in the pastoral system. Larger herd sizes reflect the importance of livestock as bankable assets, and the value of cattle and small ruminants is more evenly distributed (i.e., cattle 1.48 billion USD, sheep 1.02 billion USD, goats 1.69 billion USD) (Table 4 and Jemberu et al. (4)). While the specialized dairy farms only hold about 3% of total cattle population, they contribute 13% of total national value (i.e., 1.59 billion USD, Table 4). This is because of their higher milk yields, driven by the dominance of crossbred and exotic breeds, and better nutritional and health management compared to cattle in the other systems.

Our results also highlight the large variation in reported values for livestock production in Ethiopia. Inconsistent values from different data sources were found when triangulating values for biomass and economic value estimation. For example, different values of the live body weight of Ethiopian cattle can be found in the FAOSTAT, the Domestic Animal Diversity Information System¹ and literature (13, 31). In this study, we combined meta-analysis and triangulating values from different sources to estimate parameters for biomass and value estimation. The population values were from national statistics which is considered as the most reliable data. However, other parameters such as sex/age specific mortality was not included in the national surveys. It is challenging to determine which values to use when different values from different sources were observed. A systematic assessment of data reliability should be considered when estimating the biomass and economic values. Part of the reason for different values from literature is that the studies were conducted at different levels (e.g., zonal, regional or national) or in different areas or years in the country. Furthermore, different definitions of one variable may exist between various data sources. For example, the age groups of cattle were defined as juvenile, subadult and adult groups in the national Agricultural Sampling Survey, while in the Livestock Market Information System, the age groups were young, immature and mature (9, 23). These definitions need to be carefully checked to see if their definitions match with each other, and good knowledge of local livestock production and official data reporting system is needed for the checking. Clear and transparent definitions of terms from different data recording systems are beneficial. Establishing an ontology of livestock disease burden analysis would reduce obstacles in triangulating variable values for livestock biomass and value estimation.

Uncertainty in biomass estimation was addressed using Monte Carlo simulation in this study. The variation in a variable can be due to the nature of the data generation. Many of the values had been produced by sampling, and they were along with confidence intervals. Sensitivity analysis offers insights into which variables are crucial for the accuracy of biomass estimation. The variables with larger impacts on the variation of the interested values need to be as accurate as possible. Likewise, the estimation of other interested values, such as economic values, would also benefit from a sensitivity analysis. This procedure will save time and resources when quick livestock biomass and value estimation are needed.

Finally, our study also reveals substantial data gaps, especially for the specialized production systems. These gaps include missing data, data which is only available at coarse scales and cannot be disaggregated to production systems, and data that is out of date. Breed structures of production systems were unknown, which reduces accuracy of average liveweight, population biomass and asset value of livestock. Age/sex specific values for farm gate prices were rarely available. The market prices of beef and milk in 2021 in the Ethiopian Livestock Market Information System (23) were used in this study, which may overestimate the output values of cattle. The hire price of an ox in literature was used to calculate the value of draft power, and it may underestimate the real value of it as the price was of 2014. To address these data gaps, assumptions were made in the calculation, which may introduce a bias. For example, equal mortality from male and female calves was assumed when sex-specific mortality was not available.

The CLM production system contributed much more biomass and output values than the other systems. The herd structures, reproductivity, productivity and value of outputs of the cattle production systems are different. Values from the national statistics, literature and other data sources are able to support the estimation of cattle biomass and value by production systems in this country. However, the variations of the variables would introduce uncertainty in the estimations. The findings of this study show that the mortality, productivity and outputs vary between cattle production systems in Ethiopia. Analyzing animal health burden by production system is therefore necessary to ensure stakeholders have sufficient data for advocacy and decision making around what investments are needed, by whom and where, for most efficient improvements in animal health and production.

The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding author.

YL contributed to the conception and design of the study and writing of the first draft of the manuscript. DM, MH, and JR contributed to the conception and design of the study. YL, WJ, and TK-J contributed to acquisition of data, statistical analysis, and interpretation of the data. DM, GC, and PS contributed to revising the manuscript. All authors have read and approved the submitted version of the manuscript.