Management and production levels of cross-bred dairy cattle in Dar es Salaam and Morogoro urban and peri urban areas

K A Gillah, G C Kifaro* and J Madsen**

Ministry of Livestock Development and Fisheries P.O. Box 9152, Dar es Salaam Tanzania
kagillah@yahoo.com
* Department of Animal Science and Production, Sokoine University of Agriculture, P.O. Box 3004, Morogoro Tanzania
** Department of Large Ruminants, University of Copenhagen, Denmark

Abstract

A study was carried out to describe management and production levels of crossbred dairy cattle in urban and peri urban areas of Dar es Salaam city and Morogoro town in Tanzania. A cross sectional study design using a random sample of 153 dairy farmers was used to collect data on management and production levels of dairy cattle. Cattle sheds were measured to establish stocking densities.

One third of farmers practised free grazing system. Poor quality forage and energy rich concentrate were the main feed resources. Farmers in Dar es Salaam and Morogoro covered 14.7±1.2 and 3.07±1.2 km/day, respectively in search of forages. Steaming up of dry cows was more common in Dar es Salaam than in Morogoro. Natural service was the predominant (P≤0.05) breeding method. Over half of cattle sheds had high stocking density and each shed accommodated 1.37 animals per 6.7m² standard space required per dairy cattle. Milk production in Dar es Salaam was higher than in Morogoro. One third of the farmers planned to decrease their cattle herds due to shortage of feeds. Therefore, addressing some of these management aspects should improve dairy cattle productivity.

Key words: cattle shed, feed, free grazing, milk production, natural service, steaming up, stocking density

Introduction

In Tanzania, dairying is one of the fast growing enterprises in the livestock sector contributing 30% of the livestock gross domestic product (GDP; URT 2012). Urban and peri-urban dairy farming has expanded considerably during the past decades as a result of inadequate milk marketing infrastructure, need for civil servants to supplement their income and the high price of raw milk in urban centres relative to the price in rural areas (Kurwijila 2001).People from different socio-economic and cultural groups are involved in dairying as a source of income, food and employment (Ishagi et al 2002, Mlozi 2005). However, the situation in Dar es Salaam is different whereby dairy cattle are kept exclusively by medium and high income people in peri urban and low density settlement areas (Jacobi et al 2000). Despite the enormous contribution of dairying in peri urban areas of Tanzania to GDP, the sub sector is constrained by factors such as genetic make-up of animals and management/ environmental factors.

Poor milk marketing is very common in Tanzania and was given the highest rank among the constraints affecting milk production in Tanga Tanzania (Swai et al 2005). Informal milk marketing dominates the marketing of milk in Tanzania and according to MoAC/SUA/ILRI (1998), over 90% of the milk marketed informally is sold as raw milk. Informal milk marketing poses health risks to consumers of raw milk as it contains appreciable number of bacteria. An example of health risks as a result of informal milk marketing was reported in Tanga Tanzania whereby 83% and 56% of all milk handled and marketed in informal markets respectively had higher coliform counts than the recommended values of less than 50000 cfu/ml and were Brucella milk ring test positive (Swai and Schoonman 2011).

Water is a very important nutrient for the dairy cows in order to perform proper physiological functions of the body. Clean and safe water is not readily available in urban and peri urban areas of Tanzania. For example, water supply service coverage in Dar es Salaam is 68% and the situation is worse in other urban areas (URT 2010). Shortage of water is critical especially during the dry season resulting into reduced milk production and health status of livestock (Guendel and Richards 2002).

Environmental contamination with heavy metals from industries and improper disposal of animal wastes poses serious health threats to urban consumers of animal products and by products. For instance, the levels of lead (3.5 mg/kg) and copper (3.6 mg/kg) contamination in the blood of cattle in Morogoro Tanzania have been reported even though the levels have not reached an alarming point (Mlay and Mgumia, 2008). Improper waste disposal has been reported in urban areas of Mbeya, Morogoro and Dar es Salaam and dumping of wastes was more common among elder household heads (22%) than among the young (9%) (Mlozi 2005). More recently, Lupindu et al (2012) in their study in Morogoro urban and peri urban areas revealed high risk of contracting zoonotic diseases as a result of improper handling of animal manure.

Farm and ranch land is inadequate in most peri urban areas of Tanzania and still more of it is turned into living quarters. In view of this, Dar es Salaam and Dodoma have developed master plans showing agriculture as a land use, and Morogoro municipal has set aside land as a green belt. Shortage of land makes urban dairy producers face difficulties in disposing animal wastes and has forced them to keep their cattle within their own residential compounds. Keeping animals close to people increases the chances of transferring zoonotic diseases such as brucellosis and tuberculosis. Zoonotic diseases are a serious concern in urban and peri urban areas of Tanzania. For instance a study conducted in Tanga, Tanzania by Cole et al (2008) reported overall prevalence of brucellosis to be 5.5 percent. Likewise, scarcity of land in peri urban areas made most dairy farmers fail to establish pastures which in turn endangers the sustainability and the successful development of urban and peri urban dairy production systems (Kavana and Msangi 2005).

Natural pasture species in the communally owned land have been the main source of feed for dairy cattle (Orodho 2006), although smallholder dairy farmers do not have full control of it. Availability of forage in terms of quantity and quality, especially in the dry season, is a major concern for dairy production in Tanzania. During the dry season, the quality of the forage is very poor and leads to low milk production (Kavana and Msangi 2005).

Studies on milk demand in Tanzania indicated higher levels of per capita milk consumption in urban centres (44 l/annum) than in rural areas (30 l/annum) and that consumption of milk is higher in lower age groups (8 years and below) than other age groups (MoAC/SUA/ILRI 1998). Higher milk demand in urban centres provides an opportunity for the dairy sub sector to expand. Unfortunately, demand for milk and milk products are much higher than actual milk production.

The above challenges have led to the development of cross bred dairy cows in Tanzania to produce milk of about 1800 l/cow/annum below the genetically possible potential yields of 2500 l/cow/annum (Msangi et al 2005, Swai et al 2005). Smallholder dairy farmers in rural areas of Tanzania manage their animals differently depending on the challenges they face and available resource bases they have. However, such management information is very limited in urban and peri-urban areas of Tanzania (Guendel and Richards 2002). Knowledge on the current management and production levels of dairy cattle may assist in improving their production through formulation of policies and enforcement of urban by laws related to keeping of animals.

In order to inform efforts to improve dairy cattle productivity in urban and peri urban areas of Tanzania, it was ideal to collect important information on farmer’s perception on the major challenges to high dairy cattle performance and characterise management practices in the study areas. Therefore, for comparative purposes, smallholder dairy farms in urban and peri urban areas of Dar es Salaam and Morogoro were randomly selected as an input to the on-going study on reproductive and productive performance of crossbred dairy cattle in urban and peri urban areas of Tanzania. The objective of this study was to describe the management and production levels of crossbred dairy cattle in Dar es Salaam and Morogoro.

Materials and methods

Study areas and animals

This study was carried out in urban and peri-urban areas of Dar es Salaam and Morogoro, both located within the same agro Ecological Zone 2 with rainfall between 500-1000 mm (URT 2007b). According to UN-Habitat (2009), Dar es Salaam is the most urbanized region in Tanzania with 93.9% of its population being urban while Morogoro is the third region in the country to have large proportion of its urban population (27.0%) above the national figure (23%) of urbanization. Dar es Salaam is located between longitudes 37⁰10” to 39⁰30”E and latitudes 06⁰15” to 07⁰40”S. Morogoro region is located between latitudes 5^o 58"and 10^o 0"S and longitudes 35^o 25"and 35^o 30"E (URT 2007b). Dar es Salaam has three Municipalities, namely Kinondoni, Ilala and Temeke but, only Kinondoni and Ilala were studied. All wards keeping cattle in Morogoro town were selected for the study.

In this study, an urban area is defined as part of town up to 10 km away from the town centre with high density areas (plots are not more than 200m²), while a peri urban area is part of town 10 km up to 30 km from town centre. This part of town has low density areas (plot sizes range between 1,200m² and 2,000m²) and some basic social services such as piped water and electricity may be missing. Both urban and peri urban areas are within the jurisdiction of a Local Government Authority (Urban Authorities Act 1982). According to urban authority bylaws, keeping of animals in urban areas of Tanzania are allowed provided they are kept under zero grazing condition and not more than four head per household (URT 2007b).The choice of the two research sites was based on the fact that they had similar agro ecological zones and have fast growing human and cattle populations.

Dairy cattle considered in this study were crossbreds between Bos taurus (Friesian, Ayrshire and Jersey) and zebu (Bos indicus) breeds. All lactating cows were hand milked two times a day. Dairy cattle were either zero grazed (stall fed) where forages are cut and carried to cattle to be fed under confinement, semi zero or free grazed. Natural pasture formed the major source of feeds while concentrate was used to supplement lactating cows during milking time.

Study design and data collection

A cross sectional study design was conducted between April and August 2010 using a random sample of 71 and 82 smallholder dairy farmers from Dar es Salaam and Morogoro, respectively. For the purpose of this study, a smallholder farm was defined as the one having 1<n<50 dairy cattle of all ages and sexes. However, for easy discussion, farmers were subdivided into three sub groups according to herd sizes namely: 1-10, 11-20 and 21-50 animals. Interviews with dairy cattle farmers using semi-structured questionnaires, focus group discussions with livestock stakeholders and personal observations using a check list were the main tools used to gather information on farmers’ demographic characteristics, types of farm labour, land size, livestock inventory/herd structure, dairy cattle feeding, milk production, processing and marketing, grazing systems and breeding methods. The hygienic condition of cattle sheds was visually observed and assessed as good/clean or poor/dirty. Cattle sheds which were cleaned with water regularly and removed manure by scribing only were assessed as being in good/clean and poor/dirty sheds, respectively. Cattle sheds measurements were taken by using a tape measure to get total cattle shed area in square metres. The stocking density was determined by computing the number of cows kept per cattle shade area.

Data analysis

Microsoft Excel was used to store and draw graphs. Descriptive statistics of General Linear Model (GLM) of SAS version 9.1 (2002) was used to describe herd size and composition, milk yield, land size and amount of milk processed. Differences in sample size among continuous variables were determined by using GLM procedures (sum of squares type III). Separation of the least square means (LSMeans) was performed by using PDIFF procedure. Chi-square test (PROC FREQ) of SAS (2002) was used to analyse associations between possible combinations of categorical variables (demographic characteristics, types of farm labour, sources of capital) and values of P≤0.05 were considered as significant.

Results and discussion

Demographic characteristics of dairy farmers

On average for both locations, nearly half of the dairy farmers were self-employed while retired officers and government employees were few (Table 1). By contrast, government employees were the most numerous in Morogoro town compared to Dar es Salaam. Similar proportions of dairy farmers with college and secondary education were observed in both study areas (Table 1). Farmers had similar levels of education although Dar es Salaam had fewer and more farmers with primary and college education, respectively, compared to Morogoro. Involvement of dairy farmers with different social economic and cultural backgrounds was reported from different studies in urban and peri urban areas of East African cities (Gillah et al 2012). Different stakeholders were involved in dairy activities to supplement their income, food and employment (Kurwijila 2001, Ishagi et al 2002, Mlozi 2005).

Dairy farmers had various sources of funds/capital for establishing and running dairy enterprises. However, the main and least sources of funding were own capital and in-kind assistance (heifer in trust) from Heifer Project International, respectively (Figure 1). Dairy farmers who applied for loans comprised of 15.4% of the total in Dar es Salaam and very few (2.8%) in Morogoro. There was a positive relationship between herd size and desire of the farmer to apply for loan; 9.7% of dairy farmers with small herd sizes (1-10) applied for loans while 4.0% (n=7) of dairy farmers with herd sizes of 21-50 cattle applied for loan. Most dairy farmers use their own capital for establishing and running dairy cattle enterprises. This practice has also been reported in Dar es Salaam (Lupala 2002) and Addis Ababa (Kassa 2003; Shiferaw et al 2003). Availability of credit services for dairy related activities is difficult. For instance, Lupala (2002) in Dar es Salaam and Ayenew et al (2009) in Bahir Dar, Ethiopia reported 4% and 16% of dairy farmers respectively to have received credit services for dairy related activities. In view of this, Baltenweck and Staal (2000) pointed out that access to credit cannot be excluded as a reason for delaying adoption of dairy cow related innovations.

Figure 1. Percentage of the farmers who used different sources of capital for establishing dairy enterprises in Dar es Salaam (DSM) and Morogoro

Feeds, feeding, breeding methods and sources of dairy farm labour

Dairy farmers depended mostly on natural pasture as feed source obtained from open spaces/communal grazing land. Forage vending is very common in urban areas where young boys cut forage from open areas/fallow land and sell it to dairy farmers. Most of the forage transactions were made along the highways. The naturally occurring green feeds in the study areas were of moderate quality, having only 6.8% crude protein and 55.6% of digestible organic matter in dry matter according to Kavana and Msangi (2005). According to NRC (1989), the available feed resources are unable to meet the crude protein requirements of a dairy cow (12% CP) producing at least 10 litres of milk per day. Kavana and Msangi (2005) reported that the feeding regime practiced by smallholder dairy farmers in peri urban areas of Kibaha and Morogoro enabled exploitation of less than 83% of the animals' potential for milk production.. Contrary to this, a good number (78.5%) of dairy farmers in Tanga municipality obtained animal feeds from farm established fodder units (Swai et al 2005). This was made possible because for quite a long time, Dutch funded dairy project in Tanga gave dairy farmers training in dairy husbandry and in calf heifers on conditions that they have to establish pasture/fodder and construct cattle sheds.

Natural pasture was collected from various sources using different means of transport. Generally, dairy farmers in the study areas used mainly vehicles and bicycles/heads to collect forage from the sources to their respective dairy units. Nevertheless, a few dairy farmers freely grazed their cattle (Figure 2). Similar results were reported by Prain et al (2010) in Nakuru, Kenya where urban dairy farmers transported their fodder using mostly bicycles. Contrary to this observation, more than half of the dairy farmers in the urban centre of Addis Ababa used donkeys to bring feeds to their farms (Shiferaw et al2003). It is worth to note that donkey transport is a most important occupation and primary income source of urban household heads in Addis Ababa (Guendel and Richards 2002).

Figure 2. Percentage of the farmers who used different means to collect forage in Dar es Salaam (DSM) and Morogoro

Dar es Salaam dairy farmers covered longer distance in search of forages relative to Morogoro (Table 2), hence it was logical for them to use mostly vehicles (hired/own) to collect animal feeds. This situation may explain the high cost of milk production (188 Tsh per kg of milk) in Kibaha (closer to Dar es Salaam) relative to 171 Tsh per kg of milk (1 USD=1, 600 Tsh) calculated in Morogoro (Kavana and Msangi 2005). Generally, transport costs are too high for smallholder farmers to access feed resources and also limit selling of manure to peri urban areas for crop production (Guendel and Richards 2002).

Availability of land for dairy activities was a major concern in urban and peri urban areas, although the size differs both within and between towns/cities (Table 2). Generally, peri urban areas had large plot/farm size compared to urban settings. However, the finding that peri urban areas had larger plots than urban areas was different to Swai et al (2005) who reported larger plot/farm sizes in urban (5 ha) than in peri urban (4.2 ha) dairy units of Tanga, Tanzania. Small land availability limits dairy farmers to expand their dairy units and face difficulties in disposing animal wastes (Lupindu et al 2012).

Very few dairy farmers in Dar es Salaam and Morogoro peri urban areas allocated land for cultivated pasture production and/or set aside grazing area (Table 2). Nevertheless, peri urban Dar es Salaam dairy farmers set aside comparatively larger land for pasture production/grazing area than farmers in Morogoro. Failure to establish pasture in peri urban dairy units was reported in Addis Ababa city (Ayenew et al 2009) but different in Tanga (Swai et al 2005). Availability of alternative sources of pasture, land scarcity, lack of pasture seeds and technical know-how on pasture establishment could be the possible reasons for not allocating land for pasture (Kavana and Msangi 2005, Mtengeti et al 2008).

Energy rich concentrate was used to supplement lactating dairy cows during milking sessions (Figure 3). According to Mtengeti et al (2008), lack of protein supplement could be due to scarcity and/or high price. The practice of giving extra concentrate to dry cows in the last trimester was more common in Dar es Salaam than in Morogoro. Furthermore, steaming up of dry cows was more common with small herd sizes (1-10 animals) than in large herds (21-50 animals). The fact that dairy farmers rarely supplement dry and early postpartum cows is in agreement with other researches in urban and peri urban areas of East African cities (Mellau et al 2009; Gillah et al 2012).

About one third of dairy farmers still practice a combination of semi zero and free grazing systems in communal grazing land. The two grazing systems were predominant in Dar es Salaam compared to Morogoro town (Table 3). The communal lands are public-owned lands that include river banks, dam edges, flood plains and fallow lands and these form the major source of dairy animals' feed in urban and peri urban areas of Tanzania. Semi zero and free grazing systems are very common practices in most peri urban areas of African cities (Gillah et al 2012). In semi zero grazing system, animals are allowed to graze on open spaces for some hours and confined during the night where they are given supplementary feeds. The fact that semi zero and free grazing systems are dominant in peri urban areas is supported by Ngigi (2005) and Mustafa et al (2011). Free grazing of livestock in Tanzanian towns is not allowed and contravenes the Municipal council by-laws which require all livestock to be kept under zero grazing system. Therefore, allowing free grazing of dairy cattle implies that possibly town council by-laws are not seriously enforced (Mlozi 2005).

Herd size influenced the choice of transport used to collect forage from outside sources. Dairy farmers with small herds used mostly bicycle/head whereas farmers with larger herds had two alternatives, either grazing or used vehicles for collecting forage (Figure 4).

Figure 4. Percentage of farmers who used different means to collect forage according to the size of herd

The methods used to breed dairy cattle varied and most dairy farmers used natural service rather than artificial insemination (Table 3). The proportion of dairy farmers who used artificial insemination was less in Morogoro than in Dar es Salaam. Fewer dairy farmers used a combination of both natural and artificial insemination in Morogoro compared to Dar es Salaam. Dairy farmers preferred natural service as opposed to artificial insemination service as earlier reported in most urban and peri urban dairy units of East Africa (Gillah et al 2012). Unreliable artificial insemination services, lower conception rates and high cost of the service scared most smallholder dairy farmers from using the service (Msangi et al 2005). Contrary to this observation, 69.7% of urban smallholder dairy farmers of Dire Dawa, Ethiopia used artificial insemination service. Possible reasons for depending on artificial insemination service in some urban areas were lack of space to keep bulls and shortage of feed (Mureda and Zeleke 2008). Nevertheless, the decision on the types of breeding methods a dairy farmer should use was not based on herd size and had no relationship with levels of education.

On average for both areas, dairy farmers depended largely on hired labour followed by a combination of hired and family labour; few farmers depended on family labour alone. A combination of hired and family labour and hired labour was more used in Dar es Salaam and Morogoro, respectively to perform most of the work in the dairy units. The types of farm labour had a bearing on the herd size, whereby small herd sizes were managed mostly by hired labour, but larger herd sizes were cared for by a combination of hired and family labour (Figure 5). According to Lupala (2002), 56% of livestock keepers in Dar es Salaam employed hired labour. The use of hired labour in performing dairy activities is common in urban cities of East Africa (Gillah et al 2012). This implies that dairy farming in urban and peri urban areas is a viable enterprise, creates employment opportunities to urban dwellers and owners of dairy cattle in those cities can afford to pay labour wages (Nugent 2000, Mlozi 2005, Orodho 2006).

The involvement of hired labour in urban dairy units was probably linked to the fact that some dairy unit owners were government employees (Table 1) and others run small businesses (Mlozi 2005). However, Salem et al (2006) gave caution on the reliance of hired labour with less dairying skills and not motivated by pointing out that it may result in mismanagement practices of the dairy unit operations. Meanwhile, family labour was more pronounced in Bishoftu, Ethiopia (Megersa et al 2011) and Kampala (Ishagi et al 2002) and accounted for 54 and 52%, respectively of the households keeping dairy cattle. This shows that some urban dairy owners cannot afford to pay labour wages since they keep cattle on a subsistence basis (Ishagi et al 2002).

Figure 5. Percentage of the farmers who used different types of farm labour according to herd size in Dar es Salaam and Morogoro

Cattle shed hygienic condition and conformity to standard design

Averaged for both locations, most cattle sheds were in good condition and there was no difference in hygienic condition between the two study areas (Table 4). Similar observations were made in several urban and peri urban dairy units of East Africa (Gillah et al 2012). Shortage of water especially during the dry season and keeping a large number of animals per unit area were some of the factors which contributed to poor hygiene of cattle sheds. Shortage of water was a common problem in most urban and peri urban areas of Tanzania (Lupala 2002, URT 2010). The majority of the cattle sheds at both locations had concrete floors; very few had earthen floors. The types of floor observed in Dar es Salaam were similar to the ones found in Morogoro. Herd size was related to hygienic condition of the cattle sheds whereby in small herds, which were mainly zero grazed, the sheds were in poorer hygienic condition than when herd sizes were larger (semi and free grazed cattle).

Cattle shed design was similar at both locations and the majority of them did not conform to recommended standards (FAO 1998). For both locations, the majority of cattle sheds did not have provision for separate sleeping and walking/feeding spaces. This observation is supported by other research findings in urban and peri urban dairy units of East Africa cities (Gillah et al 2012). Poor cattle shed design decreases the productivity of the dairy cows (Kassa 2003) and predisposes them to body injuries and diseases (Aleri et al 2011). The poorly designed cattle sheds implies that they were constructed without taking into consideration the space requirement per animal as stipulated in the Tanzania Animal Welfare Act (URT 2008). The few cattle sheds that conformed to standard design were a result of a donor-funded dairy project (Heifer Project International) which required the dairy farmer to construct a standard cattle shed before starting the project. Generally, cattle sheds in smallholder dairy units were small structures made of cement blocks, burnt bricks or pieces of timber off-cuts. They are normally covered by corrugated metal sheets or thatched with grasses. The floors were made of stones or concrete.

Stocking density, herd size and structure

More than half of cattle sheds had high stocking density (Table 5) than the recommended space requirement of one animal in 6.7m² (FAO 1998). However, the stocking densities in the study areas were not different. High stocking density was a problem in urban dairy units of Tanga Tanzania (Shirima and Msanga 2004) and Nairobi, Kenya (Aleri et al 2011). High stocking density affects animal performance, and according to Moore (2010) dairy cows should not be housed at a stocking density of more than 1.17 cows per stall, since the lactating cows may produce low milk yield and have poor conception rates. Therefore, dairy farmers should be trained and sensitized to have cattle sheds that conform to standard design in order to avoid the negative effects of having smaller cattle shed per animal.

The current average dairy herd size (Table 5) in Dar es Salaam is larger than has been reported in the previous studies in urban and peri urban areas of Tanzanian towns/cities (Gillah et al 2012). The two study areas had more than the recommended herd size of keeping not more than four animals per farmer (Urban Authorities Act 1982). For long time, farmers in urban areas of Tanzania have not reduced their herd sizes despite the by-laws enacted in all urban centres concerning keeping livestock in towns. The need for more income (Gillah et al 2012) and inefficiency in by-laws enforcement mechanisms (Mlozi 2005) could possibly led to many dairy farmers to contravene the urban by laws.

Milk production and processing

The average milk yields in the study areas were different (Table 6), but within the range obtained from other earlier studies in urban and peri urban dairy units of East African cities (Gillah et al 2012). The difference in milk yield between the two study areas could possibly be explained by the fact that dairy farmers in Dar es Salaam used more artificial insemination (Table 3) and steamed up dry cows in the last trimester (Figure 3) than dairy farmers in Morogoro. Moreover, greater prices paid for a litre of milk (Table 6) in Dar es Salaam compared to Morogoro gave them an incentive to intensify management to produce more milk. Nevertheless, the average milk production in the two study areas (8.3 litre/cow/day) was lower for dairy cows with genetic potential of producing at least 15 litres of milk per day (Msanga and Kavana 2002). Such low milk production could be due to poor nutritive value of feeds and improper feeding of lactating and dry cows which do not meet animals’ physiological requirements (Kavana and Msangi 2005).

Almost all dairy farmers sold raw milk directly to consumers and very little was processed into fermented milk products (Table 6). One of the reasons contributing to low milk processing was a small amount of milk produced per household per day relative to the demand for raw milk (Table 6). This finding concurs with Shiferaw et al (2003) who reported 61.8% of dairy farmers did not process milk because of low milk output. Selling and drinking of raw milk may transmit milk borne diseases to humans (Makita et al 2010) and illegal in Tanzania (URT 2007a). It is worth noting that the dairy inspectors in Tanzania who are charged with such duties and given powers in the performance of their functions are not performing their responsibilities.

Milk collection centres were not available and dairy farmers in Dar es Salaam were forced to cover longer distance to milk market places compared to Morogoro (Table 6). Longer distance to milk market places may have a bearing on the profit margin realized from sale of milk as well as the keeping quality of the milk (IDF, 1990).

Future plans in dairying

A slightly greater proportion of farmers preferred to increase their cattle herd size than maintaining or decreasing it (Table 7). Similar observations were made in studies carried out in urban and peri urban dairy units of Dar es Salaam (Nugent 2000) and Hue City in Vietnam (Schiere 2001).

Various reasons induced dairy farmers to make decisions on either to increase, maintain or decrease their cattle herd and differed between the two study areas (Figure 6). Shortage of feed and/or grazing land and prohibitive urban by-laws that required each dairy farmer to have not more than 4 head at a time were the first and second reasons for wanting to decrease cattle herd. Availability of feeds in terms of quality and quantity has been an important constraint in urban and peri urban areas of Addis Ababa in Ethiopia (Shiferaw et al 2003, Mureda and Zeleke 2008), Kampala in Uganda (Atukunda et al 2003) and Tanga in Tanzania (Swai et al., 2005). Due to inadequate feeding, the production potential of most crossbred cows has not been reached (Kavana and Msangi 2005). The fact that dairy farmers wanted to decrease cattle herd as a result of by-laws, implies that some farmers are becoming aware of the rules and regulations regarding dairying in urban and peri urban areas. The urge to decrease cattle herd due to external forces such as market fluctuations, lack of capital, diseases and problems of manure disposal has also been in Hue in Vietnam (Schiere 2001).

Figure 6. Percentage responses of farmers on reasons for wanting to decease cattle herd in Dar es Salaam and Morogoro towns

Conclusions

Dairying in Dar es Salaam differs significantly to Morogoro and crossbred cows produced milk below their production potential. Dairy farmers depended mainly on poor quality natural pastures. Artificial insemination service was rarely used in urban and peri urban areas because of its unreliability and high cost. Dairy cattle were confined in poorly designed sheds at high stocking densities which are against animal rights stipulated in Tanzania Animal Welfare Act. Some dairy farmers were forced to decrease their cattle herds mainly due to shortage of feed/forage and prohibitive urban by laws. In order to offset the seasonal variations in feed availability, dairy farmers should adopt strategic supplementation of dairy cows using protein rich concentrates.

Acknowledgements

The authors would like to acknowledge the DANIDA peri urban livestock farming project for financial support to this research as part of the PhD study program of the first author. We also thank the dairy farmers and ward extension staff in Morogoro and Dar es Salaam urban and peri urban areas for co-operation and provision of valuable information for this study.

References

Aleri JW, Nguhiu-Mwangi J and Mogoa E M 2011 Housing-design as a predisposing factor for injuries and poor welfare in cattle within smallholder units in peri urban areas of Nairobi, Kenya. Livestock Research for Rural Development. Volume 23, Article # 53. http://www.lrrd.org/lrrd23/3/aler23053.htm.

Atukunda G, Baseke F, David S, Jagwe J, Kalyebara M, Kaweesa M, Miiro R, Musoke P, Nabulo G, Namagembe A, Niringiye C, Nyapendi R, Odongo B, Owori C, Azuba,S M and Tumwegamire S 2003 . Farming in the city: Participatory appraisal of urban and peri-urban agriculture in Kampala, Uganda. CIAT Africa Occasional Publications Series, No. 42. CIAT, Kampala, Uganda. 44 pp.

Ayenew YA, Wurzinger M, Tegegne A and Zollitsch W 2009 Handling, processing and marketing of milk in the North western Ethiopian highlands. Livestock Research for Rural Development. Volume 21, Article # 97. http://www.lrrd.org/lrrd21/7/ayen21097.htm.

Baltenweck I. and Staal SJ 2000 Determinants of adoption of dairy cattle technology in the Kenyan highlands: a spatial and dynamic approach. Paper presented at the International Association of Agricultural Economics (IAAE) Meetings, Berlin, August 2000.

Cole DC, Lee-Smith, D and Nasinyama GW (eds) 2008 Healthy city harvests: Generating evidence to guide policy on urban agriculture, CIP/Urban Harvest, and Makerere University Press, Lima, Peru 259 pp.

FAO 1998 Farm structure in tropical climates: A Textbook for structural engineering and design Bengtsson, L.P. and Whitaker, J.H. (Editors). FAO/SIDA Cooperative Programme. Rural Structures in East and South-East Africa 167 pp. http://www.fao.org/docrep/s1250e/S1250E00.htm#Contents

Gillah K A, Kifaro G C and Madsen J 2012 Urban and peri urban dairy farming in East Africa: A review on production levels, constraints and opportunities. Livestock Research for Rural Development. Volume 24, Article #198. http://www.lrrd.org/lrrd24/11/gill24198.htm.

Guendel S and Richards W 2002 Peri -urban and urban livestock keeping in East Africa - a coping strategy for the poor? Scoping study commissioned by the Livestock Production Programme (LPP) July 2002. NRI Limited, Aylesford. 29 pp http://www.tropentag.de/2003/proceedings/ node148.html.

IDF ternational Dairy Federation 1990 Handbook on Milk collection in warm developing countries IDF, Brussels, Belgium 148 pp.

Ishagi N, Ossiya S, Aliguma L and Aisu C 2002 Urban and peri-urban livestock keeping among the poor in Kampala City. Ibaren Konsultants, Kampala, Uganda 97 pp.

Jacobi PA Mend J and Kiango S 2000 Urban agriculture in Dar es Salaam: providing for an indispensable part of the diet. In N. Bakker, M. Dubbeling, S. Guendel, U. Sabel Koschella and H. de Zeeuw eds. 2000. Growing cities, growing food, urban agriculture on the policy pp 257-283 agenda, http://www.ruaf.org/sites/default/files/DaresSalaam.PDF

Kassa TM 2003 A study on urban agriculture: the case of small-scale urban dairy farming in selected areas of Addis Ababa. A thesis presented to the school of graduate studies of the Addis Ababa University in partial fulfilment of the requirements for the degree of Master of Arts in Geography 175 pp

Kavana PY and Msangi BSJ 2005 On farm dairy cattle feeding experience in eastern zone of Tanzania Livestock Research for Rural Development Volume 17 Article # 6 http://www.lrrd.org/lrrd17/6/kava17062.htm

Kurwijila LR 2001 Evolution of dairy policies for smallholder production and marketing in Tanzania In: Rangnekar D. and Thorpe W. (eds). 2002. Smallholder dairy production and marketing, Opportunities and constraints. In: Proceedings of a South-South workshop held at NDDB, Anand, India, 13-16 March 2001. NDDB, Anand, India, and ILRI, Nairobi, Kenya. 538 pp.

Lupala A 2002 Scoping study on urban and peri-urban livestock keepers in Dar es Salaam City. University College of Lands and Architectural Studies (UCLAS), Dar es Salaam, Tanzania 71 pp.

Lupindu AM, Ngowi HA, Dalsgaard A, Olsen JE and Msoffe PLM 2012 Current manure management practices and hygiene aspects of urban and peri-urban livestock farming in Tanzania, Livestock Research for Rural Development 24 (9) 2012 http://www.lrrd.org/lrrd24/9/lupi24167.htm

Makita K, Fèvre EM, Waiswa C, Eisler MC, Welburn SC 2010 How human Brucellosis incidence in urban Kampala can be reduced most efficiently? A stochastic risk assessment of informally-marketed milkPLoS ONE 5 (12): e14188.doi:10.1371/journal.pone.0014188.

Megersa M, Ashenafi F, Anteneh W and Tariku J 2011 Herd composition and characteristics of dairy production in Bishoftu town, Ethiopia Journal of Agricultural Extension and Rural Development 3 (6):113-117.

Mellau LSB, Sarimbo AP, Kimera SI and Phiri ECJ 2009 Recumbence syndrome in cattle: A study of risk factors of milk fever in dairy units in Dar es Salaam, Tanzania Tanzania Veterinary Journal Vol. 26 (2): 56-63.

Mlay PS and Mgumia Y O 2008 Levels of lead and copper in plasma of dairy cows, pastures, soil and water from selected areas of Morogoro suburbs, Tanzania. Livestock Research for Rural Development. Volume 20, Article #60. http://www.lrrd.org/lrrd20/4/mlay20060.htm.

Mlozi MRS 2005 Urban animal agriculture: its palliativity and reasons for persistence in Tanzanian towns. RICS Foundation pp. 26 http://www.rics-foundation.org

MoAC/SUA/ILRI 1998 Rapid appraisal of the Tanzania dairy subsector: Executive summary. Volume 1. ILRI, Nairobi, Kenya.

Moore DA 2010 Stocking density: How much? Western Dairy News 10 (11):167-168. http://animalscience-extension.tamu.edu/dairy/wdn.html.

Msanga YN., and Kavana PY 2002 Performance of F1 Frisian x Boran crossbred bulls and their off-springs to Tanzanian shorthorn Zebu managed under the extensive traditional system. In E. K. Batamuzi, & A. J. P. Tarimo (Eds.), In: Proceedings of the First Collaborative Research Workshop on Food Security May 2003. TARP II SUA Project pp 196-200.

Msangi BSJ, Bryant MJ and Thorne PJ 2005 Some factors affecting variation in milk yield in crossbred dairy cows on smallholder farms in North-east Tanzania. Tropical Animal Health and Production, 37:403-412.

Mtengeti EJ, Phiri ECJH, Urio NA, Mhando DG, Mvena Z, Ryoba R, Mdegela R, Singh BR, Mo M, Wetlesen A, Lørken T and Reksen O 2008. Forage availability and its quality in the dry season on smallholder dairy farms in Tanzania, Acta Agriculturae Scandinavica, 58: (4):196-204.

Mureda E and Zeleke Z 2008 Characteristics and constraints of crossbred dairy cattle production in lowland areas of Eastern Ethiopia. Livestock Research for Rural Development. Volume 20, Article #57 http://www.lrrd.org/lrrd20/4/mure20057.htm.

Mustafa E, Murtada EE, Omer A and Amir S 2011 The contribution of dairying to household welfare of the small commercial dairy keepers in Khartoum north province (KNP), Sudan, Research opinions in Animal and Veterinary Sciences, 1(1):55-59.

Ngigi M 2005 The Case of Smallholder Dairying in Eastern Africa International Food Policy Research Institute EPT Discussion Paper 131, 93 pp.

NRC (National Research Council) 1989 Nutrient Requirements of Dairy Cattle 6^th edition. National Academy Press, Washington D.C., USA.

Nugent R 2000 The impact of urban agriculture on the household and local economies. In: N. Bakker, M. Dubelling, S. Gundel, V. Sabel-Koschella, and A. Zeeuw (eds.), Growing Cities, Growing Food: Urban Agriculture on the Policy Agenda. Feldafing, Germany: Food and Agriculture Development Centre (ZEL).

Orodho AB 2006 Intensive forage production for smallholder dairying in East Africa. In: Grasslands: Developments Opportunities Perspectives by (Reynolds & Frame eds) chapter 18 http://www.fao.org/ag/AGP/AGPC/doc/Newpub/napier/eastafrica_orodho.htm

Prain G, Karanja N and Lee-Smith D (Editors) 2010 African urban harvest: Agriculture in the cities of Cameroon, Kenya and Uganda. IDRC, 335 pp.

Salem MB, Djemali M, Kayouli C and Majdoub A 2006 A review of environmental and management factors affecting the reproductive performance of Holstein-Friesian dairy herds in Tunisia. http://www.lrrd.org/lrrd18/4.htm Livestock Research for Rural Development 18 (4) 2006

SAS (Statistical Analysis System) 2002 User Installation Guide for the SAS system version 9.1, SAS Institute Inc., Cary, North Carolina, USA 221 pp.

Schiere H 2001 Livestock production in urban conditions, problems and opportunities, concepts and methodologies. In: Proceedings of the 10^th International Conference of the Association of Institutions for Tropical Veterinary Medicine “Livestock Community and the Environment”, 20-23 August 2001 Copenhagen, Denmark.

Shiferaw Y, Tenhagn BA, Bekana M and Kassa T 2003 Reproductive performance of crossbred Dairy cows in different production systems in the central Hhighlands of Ethiopia. Tropical Animal Health and Production 25:551-61.

Shirima EJM and Msanga YN 2004 The state of livestock welfare in Tanzania: An overview. In: Proceedings of the 31^st Scientific Conference of TSAP held at Kilimanjaro Crane Hotel Moshi 5^th–7^th 2004, 31:137-146.

Swai ES, Karimuribo ED, Schoonman L, French NP, Fitzpatrick J, Kambarage D and Bryant MJ 2005 Description, socio-economic characteristics, disease managements and mortality dynamics in smallholder's dairy production system in coastal humid region of Tanga. Tanzania. Livestock Research for Rural Development 17 Article #4 16 pp.

Swai ES, Schoonman, 1. 2011 Microbial quality and associated health risks of raw milk marketed in the Tanga region of Tanzania Asian Pacific Journal of Tropical Biomedicine pp 217-222. http://www.elsevier.com/locate/apjtb.

UN-Habitat 2009 Tanzania: Morogoro profile Tanzania United Nations Human Settlements Programme Regional and Technical cooperation Division 32 pp. http://www.unhabitat.org

Urban Authorities Act 1982 Printed by the Government Printer, Dar es Salaam, Tanzania 59 pp.

URT (United Republic of Tanzania) 2007a The Dairy Industry Regulations. The Government Printers Dar es Salaam Tanzania 48 pp.

URT (United Republic of Tanzania) 2007b Morogoro regional socio economic profile,122 pp. http://www.tzonline.org/pdf/Morogoro.pdf

URT (United Republic of Tanzania) 2008 The Animal Welfare Act. No. 19 of 2008 Ministry of Livestock Development 62 pp. http://www.mifugo.go.tz

URT (United Republic of Tanzania) 2010 Water sector status report ministry of water and Irrigation 127 pp. http://www.maji.go.tz

URT (United Republic of Tanzania) 2012 National Sample Census of Agriculture, Small holder agriculture Volume III: Livestock Sector. The National Bureau of Statistics and the Office of the Chief Government Statistician, Zanzibar 187 pp.

Received 15 July 2013; accepted 9 August 2013; Published 4 September 2013

Go to top