Two general questionnaires were sent to corresponding partners from 13 countries in SSA. The countries were selected to provide sufficient representation from East, South and West Africa, and included Ethiopia, Kenya, Malawi, Rwanda, Zambia, Cameroon, Ghana, Nigeria, Senegal, Togo, Mali, Burkina Faso, and Niger. The first questionnaire consisted of questions regarding current manure management practices, stakeholders involved, and their level of engagement. Respondents answered questions for each ministry in their countries responsible for activities related to manure management. These included questions on regulations regarding stocking rates, manure storage and treatment, anaerobic digestion, manure application, air pollution, water pollution, spatial planning of farms, and zoonotic diseases. The second questionnaire captured how the political environment influences the adoption of improved manure management practices. Interviewees were selected with the support of government officials from the Ministry of Agriculture, Ministry of Environment, and through the UNFCC focal point in the respective countries and were conversant with manure management policies and practices in their respective countries. Two respondents were selected per country, each from a separate ministry or NGO, responding individually to the questionnaires. All respondents were then conveyed to a face-to-face meeting where their original responses were validated through discussions among respondents from the same country and also through interaction with respondents from different countries.

In-depth field surveys were conducted in only two countries: Ethiopia and Malawi. Two regions with higher livestock densities than the national average were selected in each country. Farmers were selected with the help of local agricultural extension workers, and covered three production scales (small, medium, and large scale). A medium scale farm was defined as a farm with 8–30 TLU (Tropical Livestock Unit = 250 kg live weight) of ruminants, or 50–200 live poultry or 25–100 live pigs. Farms with fewer animals were considered as small scale, while those with more animals were considered to be large scale. A total of 23 farmers from Ethiopia and 20 farmers from Malawi were selected. Additionally, six prominent extension workers, three from each country, and working in the selected study regions were interviewed using a semi-structured questionnaire with both open- and close-ended questions. Interviews with extension workers gave a general understanding of the study environment, farming, and extension systems, while the interviews with farmers enabled a better understanding of the livestock production systems, housing, collection, treatment, storage, and application of manure as well as farmers' opinions regarding manure management.

This research was conducted in accordance with the Netherlands Code of Conduct for Scientific Practice, with the additions and qualifications from the Wageningen UR Ethical Guidelines. An ethics approval was not required for this study because it was not using humans or animals as subjects. Respondents at national level had a written informed consent by email and had the option to accept or decline their participation in the study. Interviewed farmers did not have a written informed consent, but were given an opportunity to decide whether to, or not to participate in the survey after a brief introduction about the purpose and content of the survey. Sensitive data (i.e., individual name, phone number, farm house location) were kept separate from the generic data and only people involved with the project could access this information. This study has not been previously published and is not under consideration for publication elsewhere. Necessary permissions have been taken for all material used in this article.

In four of the countries surveyed (Kenya, Malawi, Rwanda, and Burkina Faso) there was at least one ministry responsible for all of the activities, while Cameroon, and Niger had ministries responsible for eight of the nine activities, with Cameroon lacking a responsible ministry for regulations on manure storage and Niger lacking a responsible ministry for regulations on spread of zoonotic diseases. The four activities with the fewest responsible ministries in the studied countries were (i) Zoonotic disease regulation (responsible ministries absent in Ethiopia, Ghana, Togo, Mali, and Niger), (ii) manure treatment regulation (responsible ministries absent in Ethiopia, Nigeria, Togo, and Mali), (iii) anaerobic digestion regulation (responsible ministries absent in Ethiopia, Nigeria, Senegal, and Togo), and (iv) Stocking rate regulation (responsible ministries absent in Ghana, Nigeria, and Togo).

However, even though most surveyed countries had ministries responsible for most livestock manure related activities, this does not mean that the regulations were effective. In most cases, the replies to the general questionnaire can be summarized into three main points:

On the other hand, there are initiatives supporting the construction and use of manure in smallholder biogas plants generating energy for cooking and lighting. These are widely promoted as a way to reduce deforestation, decrease labor demands for women and children and improve respiratory health of women (NBPE, 2007; ABPP, 2018). This is also the case with the SNV funded African Biogas Partnership Programme (ABPP) covering five countries: Burkina Faso, Ethiopia, Kenya, Tanzania, and Uganda (ABPP, 2018), and the Green growth and Climate resilience program in Rwanda (GGCR, 2011). Meanwhile, in Kenya, other private organizations offer services which facilitate the use of manure management to promote sustainable agriculture (SGS, 2017).

The previous paragraphs show that in SSA countries, manure management policies are not explicit enough to oblige or encourage farmers to benefit from the nutrients in manure. There are however other factors that might discourage the use of manure as a fertilizer, especially its bulky nature. In Ethiopia for example, Ketema and Bauer (2011) showed that crop farmers who could afford synthetic fertilizers had lower rates of manure use because they found manure application to be more laborious in comparison to synthetic fertilizers. This implies that policies that subsidize the cost of synthetic fertilizers may discourage the use of manure, if incentives are not made for manure use.

The results show that all farmers, regardless of farm size, were able to access training, and extension services from both government and non-government agencies (Table 2). There were differences in the level of privatization of the extension services with a dominant public sector extension in most countries, with the exception of Kenya, which has a stronger private sector extension service, and Malawi, which has a mix of both (Mulanga and Jayne, 2006; Bernahu, 2012; Simpson et al., 2012). The extension services from both the government and private sector were available for farmers of all production scales, although there was a focus on smallholder farmers in most countries. Respondents highlighted that although these services existed, manure management was rarely covered by the extension scheme. Based on case studies in the various countries, respondents revealed that most farmers who adopted improved manure management practices had seen other farmers successfully implement the adopted practices. We concluded that the most effective extension services in disseminating good manure management practices were those with demonstrations that allowed farmers to see what fellow farmers were doing. Ethiopia and Malawi used lead farmers to carry out various manure management activities such as proper collection and storage, bio-slurry fertilization, composting, and application of manure on crop farms. These farms were then used as demonstration and training sites where other farmers could appraise the difference between crops in an integrated manure management system and those without.

Subsidies also tend to be available to farmers of all production scales, although this is more often the case in Eastern Africa compared to Western Africa. In Western Africa, it was common to have agricultural subsidy programs that have been mismanaged and sometimes abandoned as was observed in Cameroon, Togo, and Niger (Fonjong, 2004; Le Magadoux et al., 2013; Mackiewicz-Houngue, 2014), suggesting that adequate agricultural financing is likely a common challenge in Western Africa. It was also noticed that Non-government organizations (NGOs) were more focused than governments in providing incentives to smallholder farms for improved manure management.

In Ethiopia, cattle are the dominant livestock species as goats are poorly adapted to the cold conditions at high altitudes and pork is not consumed by people associated with the dominant orthodox religion (Tekle et al., 2013). Farmers of the landless systems closer to major urban centers have a higher proportion of crossbred cattle with higher milk yield (Tegegne et al., 2013). Further away from the urban centers, livestock systems are mainly mixed with crop production, which is typically the major source of income. In these areas, livestock management is less intensive, with the animals grazing during the day and brought back to a compound at night. The overnight housing of animals (especially cattle) in the various confinement systems (Table 4) is not only meant to shelter the animals but, most importantly, to protect the animals from theft and predators. Most cattle owners in these district also own sheep, while poultry production is predominantly (95%) in traditional systems where chicken scavenge around households for their food (Wilson, 2010).

In Malawi, farming systems are typically a mix of annual crops, which provide most of the farm income, with a combination of different livestock species including cattle, goats, pigs, and chickens. In general though, cattle are considered by farmers to be the most important. Tobacco is a common cash crop, while forest plantations and unmanaged shrub land were common as well. Communal land, where many of the farmers take their cattle each day to graze, was also noted. The number of landless farms was very low and was limited to pigs, poultry and occasionally dairy cattle. Farms also tended to have good access to markets because government-run livestock markets were established in regions that are far from major centers.

In Malawi, most of the farms that used confinement systems that combined fences and roofs (and occasionally a concrete floor) were for pigs, poultry, and small ruminants (Figure 2). In the large laying hen operations the chickens were confined all day. Pigs and dairy cattle were also confined most, if not all, of the day. The confining of dairy cattle seems to be a result of information from the extension workers as many respondents confirmed that some extension workers put strong emphasis on the confinement of dairy cattle. However, a majority (56%) of the cattle farms that had ma/inly beef cattle only have fences without a floor and roof. Similar to Ethiopia, the beef cattle are taken out to graze in the morning and return to the pen shortly before dusk. This means that not all manure can be collected and most of it stays on the grazing land and around the homesteads.

The results of the manure management practice surveys in Ethiopia and Malawi are shown in Table 3, which shows that the use of bedding material is more common in Malawi than in Ethiopia. Less use of bedding material implies that nutrient conservation in manure will likely be reduced (see section Implications of Livestock Systems and Manure Management Practices on Manure Quality as a Fertilizer in SSA). The digestate (i.e., bio-slurry from biodigesters) can be used as a fertilizer, however only 25% of Ethiopian farms with biodigesters used the digestate as a fertilizer, while all farmers in Malawi used the digestate as a fertilizer. In Ethiopia, the large farms with biodigesters discharged most of the digestate into waterways while the smaller farms dried it to obtain dung cakes used for cooking. In Ethiopia, manure is the preferred fuel source for cooking the national food (injera) because it burns slowly with a desirable deep heat required for cooking injera. However, research is ongoing to produce stoves that can cook injera using the biogas from the digesters.

Urine storage is uncommon both in Ethiopia and Malawi, likely because the predominant confinement systems do not include an impermeable floor, and so the urine tends to infiltrate the soils underlying the kraal. This means that a huge amount of nutrients (notably nitrogen and potassium) is lost, potentially damaging the local environment. In both countries, manure is generally stored as a solid. In Ethiopia almost all farms stored and dried solid manure that was later used for fuel, while over two thirds of the farmers in Malawi stored solid manure for later use as a fertilizer. In both countries smaller and medium scale farms were more likely to use manure as a fertilizer than the larger farms because small and medium farms also tended to have cropland while large farms primarily focused on livestock with limited or no cropland. Additionally, none of the interviewed farms in Ethiopia and Malawi had waterproof floors nor roofing/cover for manure storage, exposing it to additional nutrient losses.

Scavenging animals, in particular the chickens, deposited most of their manure around the household, which was not usually collected. In the Awassa region of Ethiopia, it was common for farmers with indigenous cattle breeds to go on transhumance with their animals (including small ruminants) for up to 3 months, leaving behind the rest of the family. This means that the use of biodigesters would not be feasible here because the digester requires daily feeding with fresh manure. Also, the manure produced during the migration will be left directly on the pasture fields, serving as a fertilizer.

In Ethiopia, poor manure management is typically related to insufficient labor at the farm, particularly with liquid manure where transportation was very laborious. Some farmers though, mixed the liquid manure with crop residues and other materials to make compost, which is generally easier to transport. Technical issues (such as a lack of equipment, storage capacity, trading infrastructure, etc.) were not considered as constraints to improved manure management. About one third of the farmers in Ethiopia had invested money in improving manure management within the last 5 years and were happy with the outcome of their investments (mainly higher crop yields). These farmers mentioned that these investments mainly were the result of them being convinced by other farmers rather than by agricultural extension workers. However, exchange of information among farmers was limited to small geographical areas, narrowing the possibilities of farmers to learn from others, reducing the adoption rate of recommended manure management practices.

The agricultural extension system has not sufficiently included manure management in its training scheme and there currently is a procedural disincentive for extension workers to educate farmers on improved manure management. Extension workers are judged and promoted partly based on the volumes of synthetic fertilizer they sell. As farmers tend to buy less synthetic fertilizer when they use more manure, extension workers do not always promote field application of manure. It is recommended that the extension system revise their criteria for compensating extension workers, for example based on increased crop yield per hectare.

In Malawi, most of the farmers believed the most crucial constraint to enhanced manure management was the lack of information. They did not have sufficient knowledge on manure management practices and their potential benefits. However, farmers did recognize that manure was useful as a fertilizer and most said that it was crucial for them. Most of them mentioned that synthetic fertilizers are expensive and that the use of manure allows them to purchase less fertilizer. Like in Ethiopia, about a third of the large scale farmers in Malawi had invested time or money over the last 5 years in improving manure management. Because these farms usually don't own cropland, it would be advisable for them to prepare compost manure which is marketable and easier to transport than fresh dung. Commercial compost producers could also buy large volumes of manure from large scale farmers and process and sell it to large scale crop farmers.

The quality of manure largely depends on the quality of the feed (e.g., digestibility, protein content, etc.) and by the processing (e.g., pelleting, extrusion, steaming, ensiling, fermenting, and grinding). Pasture grasses from tropical regions including SSA usually contain more lignin compared with grass from temperate regions (Van Soest, 1994), which reduces the digestibility. In grassland based systems in SSA, particularly during the dry periods, animals consume low quality grass, and straw (Lamy et al., 2012), which has been linked to excreta (dung + urine) with low N contents (Pelster et al., 2016), reducing in a poorer quality fertilizer. The processing of the feeds also affects the digestibility and therefore the feed efficiency (Stark, 2012). In SSA, feed processing is most commonly practiced in medium and large-scale landless cattle farms where the use of concentrate feed is highest. In Kenya for example, some of these farms possess their own mechanical or electrical grass chopper, to make the feed finer facilitating nutrient absorption and hence reducing the manure volume and nutrients excreted (Kiptot et al., 2015).

Collecting a mixture of bedding material, feed waste, flushing water, feathers, soil, etc. together with animal excreta will also affect the nutrient content. Bedding can conserve nutrients in manure if it partly covers the manure and can prevent ammonia volatilization (KATC, 2004). Collecting manure with beddings can also result in manure with lower mineral N concentration and higher C:N ratio because the bedding materials (e.g., straw) usually have lower N concentration than the animal excreta (Lekasi et al., 2002). Increasing the C:N ratio will result in greater rates of N immobilization by soil microbes (USDA, 2011) reducing volatilization losses, although proper timing of application is critical as the N immobilization by microbes may reduce N availability for the growing crop. For the two surveyed areas, there were three manure categories that can be distinguished: (a) Liquid manure (slurry): collecting a mixture of feces and urine in liquid form where the animals are most often kept on slightly sloping solid floors that are regularly cleaned and flushed with water; (b) Mixed manure: comprising both solid and liquid manure streams from animals kept on bedding material. (c) Solid manure: excreta from animals collected as solids with or without bedding material (Martinez et al., 2009; Lenkaitis, 2012).

The three most common ways of treating manure in sub-Saharan Africa include drying (active and inactive), solid storage, sometimes with active composting, and anaerobic digestion.

The potential of biogas production in reducing GHG emissions has been debated by previous authors. The Netherlands Development Organization for example, states that one biogas plant has the potential to reduce 4t CO₂ equivalents of emissions per year (NBPE, 2007) and produce bio-slurry (described in Table 5) which is a rich fertilizer (Warnars and Oppendoort, 2014). On the other hand, Bruun et al. (2014) stated that biogas digesters are often poorly managed and that there is a lack of proper distribution system for biogas. This results in unintentional release of methane through leakages, and intentional flaring of the gas during surplus production periods (Bruun et al., 2014), with a risk of increasing, instead of reducing GHG emissions (Vu et al., 2015).

Several countries in SSA have biogas programs to promote the use of biogas by livestock farmers (ABPP, 2018). However, the adoption of biogas technology by livestock farmers in SSA has been limited mainly due to high investment costs and lack of management skills, although current programs usually have some subsidies as incentives to help farmers overcome the high investment costs (NBPE, 2007; Ngigi, 2010). Ethiopia for example has a relatively high adoption rate of biogas compared to other African countries due to aggressive government policies combined with incentives including subsidization of construction costs.

There are many different manure storage systems (Lekasi et al., 2002), however in SSA, smallholder farmers tend to use solid storages with no collection or storage of urine. In the kraal systems where animals graze freely during the day and spend the nights inside the kraal, the manure is managed as a drylot system where the animals deposit much of the manure in the kraal where it is allowed to pile up before infrequent collection. In other systems where animals are confined (such as in zero-grazing), dung is collected and stored in heaps, mostly without a hard floor or cover. Liquid manure systems, such as silos, pits or lagoons are applied primarily with the anaerobic biodigesters or in conjunction with larger scale, and highly mechanized systems. Opio et al. (2013) estimated the frequency of the various cattle manure storage systems as defined by IPCC (2006), in order to calculate GHG emissions in ruminant production systems. This classification shows manure management in SSA livestock systems as shown in Figure 3, where the majority are drylot and pasture systems with limited options for integrated manure management.

In the case of pig and poultry, most of the animals are kept in backyard systems. For pigs, a wide range of manure storage systems have been applied as described by MacLeod et al. (2013), with commercial systems in SSA managing large fraction of the livestock manure in the drylot system (Figure 3). Backyard chickens, which are more common in SSA, are mostly free ranging and as a consequence, 50% of the manure is deposited randomly in the farmyard, with the other 50% deposited and applied on the field (MacLeod et al., 2013). For commercial chicken operations, the animals are confined most or all of the time and manure is stored either as deep litter (broilers) or in pits (layers). In broiler deep litter systems, manure is combined with the litter (commonly sawdust) and this combination is only removed after a complete batch of birds have been reared, typically 5–7 weeks in SSA. Figure 4 shows a classification of manure management systems for poultry and pig in SSA, based on MacLeod et al. (2013).

Emissions from stored manure are strongly related to manure characteristics (e.g., liquid or solid manure), storage conditions (temperature, wind, humidity, etc.) as well as the storage duration of manure. Lorimor (2015) showed that ammonia loss is positively correlated with temperature, humidity, and wind speed, and that urine exposed to the air at 38°C lost 92% of its N in 7 weeks while it took 12 weeks to lose the same amount of N at 32.5°C. Rufino et al. (2007) found that manure covered with a plastic film and stored with roofing lost 20% of nitrogen compared to 55% nitrogen loss in manure that was stored in open heaps in Kenya. Another study in Western Kenya also found that manure stored in open pits had lower mass fractions of N and P than manure in heaps under roof and in open heaps (Tittonell et al., 2010). As a result, manure storage was identified as a key management variable affecting the efficiency of nutrient return to the soil (Snijders et al., 2008) with application of manure stored in covered pits increasing maize yields in Zimbabwe by 104% compared to farms that stored manure in open heaps (Mutiro and Murwira, 2003). It is therefore recommended to shade manure storage facilities as much as possible in order to reduce exposure to high temperatures and subsequent N losses, as well as limiting exposure to rainfall, and thus minimizing nutrient losses due to leaching. The long term storage of liquid manure in lagoons or deep pits produces more GHG (i.e., methane), compared to the storage of solid manure, due to the anaerobic conditions in liquid manure storages. Shorter storage periods also reduce N losses (Sumberg, 1998; Snijders et al., 2013) and are also recommended for SSA.

In SSA, solid manure is commonly applied to agricultural fields in holes, in furrows or spread/broadcasted and incorporated. It is recommended that the manure be incorporated into the soil immediately after application, as it will retain more nutrients (in particular N) that will later be available to crops (Snijders et al., 2008). For example, 90% of N from liquid manure will be available to plants if it is incorporated within 8 h compared to only 40% N availability if incorporation is done after 5–7 days (Snijders et al., 2008). It is worth noting that in most cases in SSA, urine, and liquid manure are not often managed or applied to agricultural soils, but are left to (over) flow, and either end up in the soil or are washed into water bodies without any treatment (Teenstra et al., 2014).

Table 5 shows a summary of the most common manure management approaches in SSA bringing out their advantages and disadvantages. In summary, the use of manure as a fertilizer is generally beneficial to the soil and can improve crop yield. However, different strategies for manure management could lead to increased preservation of nutrients in manure, enhancing its use as a fertilizer.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.