Livestock Research for Rural Development 26 (6) 2014 Guide for preparation of papers LRRD Newsletter

Citation of this paper

Some factors associated with poor reproductive performance in smallholder dairy cows: the case of Hai and Meru districts, northern Tanzania

E S Swai, P Mollel* and A Malima**

Veterinary Investigation Centre, PO Box 1068, Arusha, Tanzania.
ESswai@gmail.com
* National Artificial Insemination Centre, PO Box 7141, Usa River, Arusha, Tanzania
** District Livestock Office, PO Box 10 Hai, Tanzania

Abstract

An on-farm observation and questionnaire based study was conducted, during the period of January to March 2009, to assess the reproductive performance of dairy cows in two smallholder dairying districts of northern Tanzania. A total of 100 (50 Meru district and 50 Hai district) smallholder dairy farms owning 1-4 pure and crossbred dairy cows were visited and the reproductive performance of 191 (94 Meru and 97 Hai) dairy cows were analysed.

 

Overall, land holdings averaged 2.26±1.99 (mean±standard deviation) acres, with an average of 0.55±0.67 acres being reserved for pasture production. Land holding and reserved land for fodder production was, on average, higher in Hai (2.62 and 0.65 versus 1.81 and 0.45 acres) compared to Meru district (P<0.05). The mean (mean ± standard deviation) number of lactating cows per farm was 1.79 ± 0.87 and ranged from 1 to 5. Survey results revealed that 40% of the smallholder dairy farms reported dairying to be their most important source of household income. Other reported sources of income were crop farming (32%) and off farm activities such as trading (12%), employment (9%) and traditional livestock keeping (6%). Perceived and reported dairy farming constraints included availability of feeds (quantity and quality) (81.8%), lack of money to buy farm inputs (77%) and insufficient land (53.0%). Others were milk marketing (31%), diseases (28%), availability of breeding bulls (27%) and high costs of inputs (18%). The birth rate was 39% and overall mean (mean ± SE) estimated calving interval (CI) was 525±18 days. Mean CI was significantly higher in Meru (530±28) than in Hai (518±22) (P<0.05). Hypothesized factors associated with long CI based on logistic regression models were body condition score (BSC) and low body weight. Cows with body condition (>3 BSC) were three fold (OR =3.8, P = 0.048) times more likely to have a reduced CI and cows with low body weight were associated with extended long CI (β for age = 0.01, P = 0.044). Despite the herdsmen having extensive dairying experience and competency in heat detection, the CI was too long and possibly associated with inadequate feeding as reflected by the low body score condition (average 2.6), low level of land holding and daily fresh matter intake per cow in most surveyed farms.

 

The present study revealed that the reproductive performance of the dairy cows, under the smallholder management conditions in the two given districts, was sub-standard which prevented attaining a calf crop every year and expected levels of milk production. Interventions should include an effective extension service to advise on improved dairy cow and feed resources management together with applied research into the factors causing extended calving intervals.

Key words: dairy cows, smallholder, reproduction performance, Tanzania


Introduction

In Tanzania, the majority (about 80%) of the 43 million strong populations depend on agriculture, mainly mixed farming. Livestock contributes about 30% of agricultural Gross Domestic Product (GDP), derived from an estimated 21.3 million heads of cattle, mostly comprised of indigenous East African short horn zebus. In Tanzania 17 million cattle are held by 1.27 million small-scale households, half of which keep 1-5 animals (FAO 2006; URT 2012); typical smallholder dairy households keep 3 animals (Swai and Karimuribo 2011). Demand for dairy products in Tanzania is driven by a growing human population (currently estimated at 43 million and growing at 3.3% annually), urbanisation (growing at 5% annually) and increasing incomes from the high economic growth rate (real GDP growth is currently about 4% per annum (MOAC/SUA/ILRI 1998). But milk supply has failed to keep pace with growth in demand. Projections suggest that, under current trends, production is very likely to fall short of demand. These trends present an important income earning opportunity for current and potential smallholder dairy producers in Tanzania and their market agents, through dairy production, processing and marketing.

 

Arusha and Kilimanjaro regions, supply about two-thirds of the milk produced in Tanzania; other significant regions are Tanga, Mwanza, Kagera and Dar es salaam, largely from smallholder dairy dairy production systems. Production in these areas in severely constrained by feed resources, including the high degree of seasonality (Nkya et al 2007). Limited quantity and quality of feed is considered to be the main reason for the low production figures of 5-10 and 0.5 litres/day for improved dairy and indigenous short horn zebu cows, respectively.

 

To sustainably improve milk production and profitability of smallholder dairy units requires an understanding and informed response to the factors constraining productivity - inclusive of reproductive performance. The main index to estimate reproductive performance is the calving interval (the period in days between two subsequent calving dates). Calving interval is an important index of cow reproductive performance and calving interval of 365 days is desirable for efficient production (Lobago et al 2006).

 

The objective of this study was to identify farm and animal-level factors affecting the calving interval (CI) of small holder dairy cows, under rural smallholder management, in the land constrained districts of Hai and Meru in northern Tanzania.


Materials and Methods

Study area

 

The study was conducted in two contrasting districts belonging to Kilimanjaro and Arusha regions. Hai is one of the six districts of the Kilimanjaro Region of Tanzania. The district lies between Longitude 37’100 E and Latitude 3 ‘100S covering an area of 1,101 km2. The district experiences both temperate (highland zone, 1600 meters above sea level, >1500mm/year, high population density), sub-humid (midland zone, 100-1600 meters above sea level, 1550 mm/year) and semi-arid climate (lowland zone, 750-1000 meters above sea level, 325 mm/year). The main ethnic groups are Chagga and Maasai pastoralists. Traditionally the Chagga have their homesteads on the highlands or mountain slopes, where, in addition to growing coffee and bananas, they also keep a few livestock, particularly cattle. The lowland farms are used mainly for the production of food crops, normally maize, beans and finger-millet. Due to land pressures the Maasai pastoralists are progressively moving from a purely cattle keeping culture to that of mixed farming and other diversified income earning activities. Administratively, the district is divided into 10 wards with a population growing at a rate of 1.9-2.7% per year (URT 2002).

 

Meru district is one of the 6 districts which form Arusha Region in northern Tanzania. The district is located between Latitude 3.50 to 3.70 S and Longitude 26.50 to 37.50E. The district experiences both sub-humid and semi-arid climate with elevation ranging from 1200-1900 meters above sea level and average annual rainfall ranging from 700 to 2000 millimeters. Rainfall is distributed into two distinct seasons, the wet season, which is extended from March to June and the dry season that runs through the remaining part of the year. The mean annual minimum and maximum temperatures are 15oC and 30oC, respectively. Agriculture, dominated by crop-livestock production system, is the main stay of the population in the district. Administratively, the district is divided into 6 divisions, composed of three major ethnic groups which are the more sedentary Wameru and Waarusha and the pastoralist Maasai.

 

Animals and management system

 

Mixed crop-livestock farming is the predominant production system in the two studied areas. The main livestock types kept in the area include cattle, sheep, goats, and poultry. Historically, dairying started during the pre independence period and dairy animals were introduced by missionaries, colonial settlers and colonial veterinary services. As a consequence of upgrading over the years, the genetic composition of the dairy stock has increased to over more than 75% exotic genes. The main livestock feed resource in the area is natural pasture, farm established pasture mainly Napier (Pennisetum purpureum), Setaria spp, Banana stem (Musa acuminata) and supplemental feeds include hay and crop residues like maize stover and bean straw. The major livestock diseases prevalent in the area include Tick borne diseases, anthrax, blackleg, brucellosis, mastitis and parasitic problems (Maleko et al 2012). Breeding of cows both in the rural and urban farms are by artificial insemination (AI) using semen from Friesian, Ayrshire, or Jersey sires. In most cases, natural mating is also used with improved cross bred bulls of unknown genetic composition. Artificial Insemination is practiced more often in Meru than in Hai due to its closeness to the National Artificial Insemination Centre (NAIC). Heat detection is commonly done visually by farmers and reported to an AI technician.

 

Study protocols

 

Farms for the study were identified from the District Livestock office database. Based on the previous studies on number of potential breadable cows per farm and the associated reproduction performance in Tanzania and elsewhere (Table 7), a sample size of 100 farms, 50 from each district were selected and enrolled for the study. The farms were visited once (cross-sectional study) during the period of January-March 2009.

 

Questionnaires survey method and data collection

 

A structured questionnaire was prepared, checked for clarity and used to collect information from dairy cow owners (each having three to five crossbred cows) kept under smallholder management conditions. The owners were interviewed and data related to reproductive performance of their dairy cows recorded. First calvers and above were the study subject of interest. Prior to the interview, respondents were briefed as to the objective of the study. To minimize bias due to farmer recall, farm operations and events captured (except weighing, aging and body score assessment) were limited to the year (2008) preceding the study. This involved detailed tracing of all animals on the farm, and examination of any written records, so that all ages of the cattle, calving dates, date of deaths and other movements of cattle on and off the farms agreed chronologically. Other information collected included details of whether or not the animal had access to minerals owner of the animals,  whether the cattle owner had attended any dairy husbandry training, the herd size, feed source and feeding regime, source and distance to breeding bull (in kilometers), awareness and monitoring of heat signs, breeding record keeping, age of the cow(retrieved from farm record or farmer recall), breed, filial generation (classified as F1, F2 and F3 based on the level of exotic genes from breeding records), source of animals (home-bred or brought-in) and body weight estimates as described by Swai et al (2007a). Body condition was scored on a 5-point scale, where 1 represents very thin and 5 represent grossly over fat, according to the guidelines described by Nicholson and Butterworth (1986).The main index of interest was the calving interval (CI). Calving interval was calculated as the difference in days between the current (most recent) calving date and the previous calving date. Some cows did not have two calving dates and a next calving date was estimated, when possible, as follows. For cows that were pregnant or bred and assumed to be pregnant, the calving date was estimated by adding 280 days to the conception date. These cows were broadly defined as ‘breed-able cows’. Other parameters investigated was birth rates(BR) defined as the proportion of total number of births in 2008 to total number of cows alive or cow days in 2008 (French et al 2001). A long calving interval (LCI) is considered to occur if the CI extends beyond the standard recommended of 430 days under tropical conditions (Mujuni 1991).

 

Data analysis

 

Data collected were edited, stored and analyzed using Epi-Info version 6.04d (Centre for Disease Control, Atlanta, USA). The unit of analysis was individual potential breedable females (cows and heifers > 30 months old) that were on the farm in 2008. The outcome (dependent) response investigated was CI as binary variable (long >430 or short <430 in days). Explanatory (independent) categorical variables investigated were history of farmer attending training, gender of animal owner, cattle breed, filial generation, source of animals, frequency of extension officer contact, source of fodder coded as binary response( yes or no). Continuous(independent)  variables (i.e. age, weight and body condition score) were transformed into (agecentre, agecentre2 , weightcentre, weightcentre2, scorecentre, scorecentre2) in order to normalize and ease data handling. Associations between dependent and independent variables were investigated in two steps by logistic regression (using Egret for Windows version 2.0, Seattle, USA) with ‘farm’ as a random effect because cows on one farm may not be statistically independent of one another (Kristula et al 1992). In the first step, relationships between each independent and outcome variable were individually investigated. In the second step, any variables that were significantly associated at the P< 0.25 level were included in multivariable regression models producing, by forwards and backwards substitution and elimination, the most parsimonious models in which all independent variables remained significant at the P < 0.05 level.


Results and Discussion

Household characteristics of participating farms

 

A total number of 100 farms (50 from 25 villages in Hai and 50 from 12 villages in Meru) participated in this investigation. Of the 98 households that participated in the study, 91 (92.9%) were headed by men  and only 7 (7.1%) were headed by women . The majority of the households were headed by monogamous married heads but seven households that were headed by females belonged to either widows (6), most of whom had been separated from their husbands for several years, or single women.

 

The study survey revealed that farmers started keeping dairy cow before the seventies (8%). The majority (81%) reported to have started between years 1970 to 2000 and eight farmers (11%) reported to have started dairy farming between years 2000 to 2009. Dairy farming experience ranged from more than 30 years for 68% and less than twenty years for 32% of the farms surveyed. The source of the first dairy animals was through cash purchase (72%) or through gift from relatives or friends (22%) and few through either heifer in trust (HIT) scheme (4%) or loan (1%) from commercial banks. Most of the animals were owned by males (89%). Nineteen (20%) of the interviewed farmers, mostly women(70%), and all from Meru, had attended and participated in some form of training on dairy husbandry or disease control courses within the last five years prior to the start of the study. Trainings, which were either partially or fully sponsored, were organized by or through Non-governmental organizations (NGOs) and government projects like Participatory Agricultural Development and Empowerment Project (PADED) etc. In addition, 28% of the farms had access to an extension agent’s visit more than two times in a year.

 

Farm demographics analysis

 

Overall, landholdings averaged 2.26±1.99 (mean ± standard deviation) acres, with an average of 0.55±0.67 acres being reserved for pasture production. Land holding and reserved land for fodder production was on average higher in Hai (2.62 and 0.65 versus 1.807 and 0.45 acres) than in Meru district (P<0.05). Smallholder dairy farms in Meru had an average herd size of 3.64 ±1.70 (mean ± standard deviation) animals, and the majority (60%) of farms had between 1-5 animals. In contrast, small holders in Hai district had an average herd size of 3.26±1.42, and number of animals ranged from 1-3. A total of 191 breed-able cows of varying age, parity, lactation stage kept on 99 farms were examined. The mean (mean ± standard deviation) number of lactating cows per farm was 1.79 ± 0.87 and ranged from 1 to 5. Details of the herd structure by age category are given in Table 1. Hai district had the lowest (P<0.05) tropical livestock units (one mature cow equivalent to 1Tropical Livestock Unit, TLU) compared to Meru district. Meru district had significantly (p<0.05) high number of immature females (<18 months old) than Hai district. Overall, the herd size did not differ significantly between the 2 districts under study (p>0.05) (Table 1).

Table 1: Average number of cattle category kept by respondents

Age

category

Average ( mean±stdDev)

Over all

(mean±stdDev)

Hai

Meru

Immature males

0.68± 0.81

0.70± 0.84

0.69± 0.81

Mature males

0.22± 0.57

0.29± 0.70

0.25± 0.60

Immature females

0.54± 0.64a

0.92± 0.69b

0.73± 0.69

Mature females

1.82± 0.92

1.76± 0.77

1.79± 0.86

Overall

3.26±1.42 a

3.64±1.70 b

3.43±1.57

a,b Means in the same row for each parameter with different superscript are different  at P < 0.05.
stdDev – standard deviation

Source of household income and their relative importance to dairying

Forty–percent of the surveyed smallholder dairy farms reported dairying to be their most important source of household income. The second most important source was crop farming (32%) and off-farm activities such as trading (12%), employment (9%) and traditional livestock keeping (6%). The interviewees ranked dairy farming above other farming activities as it provides a high and regular income. There was a significant difference between the two study districts, with dairying being ranked higher in Meru than in Hai (P<0.05) (Figure 1). Many studies (Van Munster 1997; Leslie et al 1999) complemented dairying for its immense contribution as a source of income for rural areas. Regular flow of cash, milk for household consumption and for collateral or security being the most cited reason (Leslie et al 1999).

Figure 1: Major source of household income (n= 100)
Perceived farm constraints

Of the 100 respondents interviewed, 81 (81%) reported different farm constraints occurring in their areas. Overall, the most common constraints mentioned included availability of feeds (quantity and quality) (81.8%), lack of money to buy farm inputs (77%) and lack of enough land (53.0%). Other constraints reported to occur less commonly were milk marketing (31%), diseases (28%), availability of breeding bulls (27%) and high costs of inputs (18%). When responses were stratified by respondent’s district, it was apparent that lack of feeds and enough land were of concern to all farmers in the two districts under study (Table 2). Lack of feeds was in agreement with the findings of Nkya et al (1999) and Leslie et al (1999), who reported animal feeds as a major farm constraint for zero grazed dairy cattle.

Table 2: Ranking of farm constraints by respondents in the ‘Meru’ and ‘Hai’ study participants

Constraint

Average rank (and range) assigned by respondents in:

Hai

Meru

Overall

Mean±StedDev

Mean±StedDev

Mean±StedDev

Lack of land

0.90 ± 1.2

(1-3)

1.32 ± 1.37

(1-4)

1.13 ± 1.30

(1-4)

Feed availability

1.08 ± 0.83

(1-3)

1.08 ± 0.63

(1-3)

1.08 ± 0.74

(1-3)

Bull availability

0.88 ±  1.46

(1-5)

0.70 ± 1.32

(1-3)

0.79 ± 1.32

(1-5)

Milk marketing

0.72 ± 1.08

(1-3)

0.62 ± 1.06

(1-3)

0.67 ± 1.07

(1-3)

Diseases

0.70 ± 1.05

(1-4)

0.46 ± 1.03

(1-4)

0.58 ± 1.04

(1-4)

Lack of money

0.50 ± 0.83a

(0-3)

0.28 ± 0.73b

(0-3)

0.39 ± 0.79

(0-3)

High  prices of input

0.20 ± 0.63a

(0-3)

0.50 ± 0.88b

(0-3)

0.35 ± 0.78

(0-3)

a, b Means in the same row for each parameter with different superscript are different at  p< 0.05)

Prioritization of farm constraints significantly varied between respondents in the ‘Meru’ and ‘Hai’ groups (P<0.01) as presented in Table 2. The likelihood of Hai farmers reporting lack of money to buy inputs was twofold when compared to Meru farmers.  In contrast the likelihood of Hai farmers ranking high cost of input was lower than Meru farmers. Overall, the first three constraints of priority to farmers in descending order include lack of land (1st), availability of feeds (2nd) and availability of breeding bulls (3rd).

 

Feeds and feeding

 

Feed resource potential and availability

 

The main sources of fodder for animals in the two studied areas is crop residues (98%), farm established fodder near homesteads (79%), road side natural pasture (64.6%) and conserved  feeds such as hay (38%)(Plate1a,b,c). In both sites, crop residues (maize stover or bean straw) were used more often during dry season period and the differences between the two sites was not significant (P>0.05). The use of farm established fodder was significantly higher in Hai (92%) than in Meru (66%) (Table 3). This implies that most farmers depend on fodder from outside as either cut fodder and/or crop residues.

Table 3: Sources of fodder for farmers in Meru and Hai districts as of 2008

Source of fodder 2008a

Number of respondents, %, (n)

Overall, %, (n)

‘Hai’ (n=50)

‘Meru’ (n=50)

Fodder plot at farmb

92 (46)

66 (33)

79 (79)

Cut fodder from outside

67.3 (33)

62 (31)

64.6 (64)

Bought fodder

28.6(14)

48(24)

38.4(38)

Crop residues

98 (48)

100 (50)

98 (98)

aIn order to minimise recall bias, respondents were asked to mention sources of their fodder for one immediate past year before investigation when they could remember many events that happened during this period

b Banana leaves and stem are considered as part of the farm established animal feed


Plate 1a: A typical size of one bundle Plate 1b: Maize stover pile–main dry season feed


Plate 1c: Farm established fodder Plate 1d: A typical stall fed cow

Out of 100 farmers interviewed, 52 (52%) reported to have a pasture plot in the year 2008(Plate1a) with an average size of 0.52±0.64 (range= 0.25 to 3) acres per farm, consistent with previous reports (Mlambiti et al 1982). It was found that the average size of pasture plot for ‘Meru’ farms was smaller (0.39±0.61 acres per farm) than the ‘Hai’ farm plots (0.65±0.32 acres per farm) (P<0.05). Other plant species established were Rhodes grass (5.3%), ‘luceana’ (5.3%) and leguminous plants (15%) mainly Caliandra and Sesbania spp.

 

Feed requirement for a dairy cow

 

As available feeds (grown at home and collected from communal land/roadside) vary so widely in quantity and quality it was difficult to establish a basic feeding standard for the dairy cattle in the study areas. Obtaining sufficient dry matter (DM), because of low dry matter content of the forage, would be challenging but it may be assumed that for the major part of the year animals were more likely to have difficulties in obtaining a sufficient supply of energy and crude protein. Under this scenario, it’s likely that the available fodder source will provide only maintenance requirement for most part of the year. Critical analysis of pasture was not done in this study, however, quantitative estimates of forage offered to cattle per day was approximately 2 bundles per one livestock unit (Plate a). (One bundle estimated to weigh 15-20 kg, One mature cow = I LU). Assuming a 10 % wastage, the actual intake was 27-36 kg per day per cow, a figure which is nearly less by 50% of the actual recommended 50-60 kg (fresh matter) intake per day per cow (Doto et al 2004). This intake is capable of meeting only 2.7-3.6 kg of dry matter against the actual recommended 7-10 kg of DM per day for a cow of 350-450 kg live weight (water content of green fodder estimated to be 90%). In addition to the low DM intake, the quality of natural grasses and crop residues such as maize stover is questionable in terms of crude protein (CP), digestibility and mineral content (Mtengeti et al 2008).  By inference, there is a severe under-feeding of dairy cows in the two study districts. Moreover, 8% of the surveyed farmers were not feeding minerals and 6% were using mineral that cannot adequately provide the required calcium and phosphorous needs for dairy cattle. It was difficult to ascertain the precise level of mineral offered to cattle in this survey.

 

Animal breeding related constraints

 

Breeding related constraints were identified and ranked  to be poor include: heat detection(58% out of 98 respondents), cows not showing regular heat signs( 46% out of 98 respondents) and high breeding cost( 42% out of 98 respondents). Others were breeding bulls being far away (10%) (>4 km from cow source) were mentioned by 10% of the respondents (Table 4). Breeding cost and distance from the nearby bull box or / source were significantly ranked high in Hai compared to Meru district (P<0.05). Most of the study farms practiced zero grazing (92; 93%) and AI services were available to 53% of the farmers. There was no significant difference with respect to the cattle rearing and breeding system between the 2 study sites (P>0.05).

Table 4: Ranking of farm breeding constraints by respondents in the ‘Meru’ and ‘Hai’ study participants

Constraint

Average rank assigned by respondents in:

Hai

Meru

Overall

Mean ± StedDev

Mean ± StedDev

Mean ± StedDev

Bull availability

0.24 ± 0.52

0.12 ± 0.33

0.18 ± 0.43

Poor heat detection

0.69 ± 0.79

0.68 ± 0.55

0.68 ± 0.68

Cow not showing heat

0.98 ±  0.97

0.96 ± 0.96

0.97 ± 0.96

High breeding cost

1.00 ± 1.02a

0.64 ± 1.02b

0.82 ± 1.12

Bull far away

0.36 ± 0.88a

0.08 ± 0.44b

0.22 ± 0.71

a,b Means in the same row for each parameter with different superscript are different at p< 0.05)

Competencies on heat detection and monitoring

Of the interviewed farms, 92% of the participants could recognize one of the cardinal signs of heat namely discharge of white mucus. Over 70 % could recognize other signs like restless, bellowing and mounting other cows (Figure 2). Other important heat signs like drop in milk yield, standing to be mounted by other cows and off-feed were less mentioned. There was no statistical significant difference with respect to the ability to recognize heat signs between the two study sites (P>0.05).

Figure 2: Awareness of heat signs (n=98)

Of the 99 farms that responded to check/monitor heat signs, 62(62.6%) and 19(19.2%) mentioned to check/monitor twice and once a day, respectively. Eighteen farmers (18.2%) reported checking heat signs once every month. A significantly greater number of farmers in Meru were keener to check heat signs than their colleagues from Hai (P = 0.029) (Table 5).

Table 5:  Frequency of monitoring heat signs

Frequency: checking

heat signs

Study sites, n, (%)

Over all

mean

n, (%)

‘Hai’

(n = 50)

‘Meru’

(n = 50)

Once every day

8 (16)

11 (22)

19 (19)

Twice a day

26 (52)a

36 (72)b

62 (62)

Once every month

16 (32)

2 (4)

18 (18)

Not checking at all

0 (0)

1 (2)

1 (1)

a,b Means in the same row for each parameter with different superscript are different at p < 0.05)

Study cows participation and characteristics

Overall, 191 potential breed-able females were examined from the two study districts (Table 6). Of these, 107(56 %) were reported to have calved more than once during their lifetime. The average number of breed-able cows per farm was 2 and ranged from 1 to 4 and the majorities were F3 with over 70% having Friesian and Aryshire blood genes. The average age of the study cows was 5.25 years with a range varying from 2.5 to 15 years. The mean (mean ± StDev) body weight (kgs) and condition score of the investigated cows was 305.82 ± 60.53 and 2.62 ± 0.55, with a range varying from 230 to 450 and 1.5 to 4.0, respectively. No single cow was recorded to have a score of 0 and >4.0. Body condition was considerably better in dry, pregnant than in heifers and lactating cows. Pregnant cows possibly due to ‘gut fill’ effect had comparatively better body score compared to other animal categories. The overall score of 2.6 is somewhat less than the recommended 3.0 for a typical well-cared cow (Van Niekerk and Louw 1990). This is clearly an indication of underfeeding possibly allied with other potential stressors such as disease, inadequate housing, poor overall management, and heat stress - a phenomenon that is common in many stall-fed dairy animals.

Table 6: The proportions of cows in each category of each variable investigated during the study (n =191)

Categories

No. of cows (%)

 

Hai
(n =97)

Meru
(n=94)

Gender of the cow owner

Male
Female

92 (94.8)
5 (5.2)

74 (78.8)
20 (21.3)

Source of animals

Bought
Home bred

37 (38.1)
60 (61.9)

39 (41.5)
55 (58.5)

Breeding

F2
F3
F4

18 (18.5)
73 (75.3)
6 (6.2)

16 (17)
78 (83.0)
0 ( 0)

Physiological status

Dry
Pregnant
Milking

10 (10.6)
35 (36)
59 (60.8)

11 (11.7)
4 (4.2)
65 (69.1)

Body score

1.5
2.0
2.5
3.0
3.5
4.0

8 (8.2)
36 (37.1)
31 (32)
19 (19.6)
2 (2.1)
1 (1.0)

0 (0)
8 (8.5)
30 (31.9)
33 (35.1)
21 (22.3)
2 (2.1)

Breed codes

Friesian cross
Aryshire cross
Jersey cross

58 (61.2)
52 (55.2)
19 (20.2)

67 (71.2)
28 (30.8)
2 (2.1

Calving interval

>430 days
<430 days

27 (27.8)
70 (72.2)

48 (51)
46 (49)

Age

3 to 5 yrs
>5 to 8 yrs
>8 yrs

64 (66)
25 (25.8)
8 (8.2)

48 (47.3)
28 (33.3)
18 (19.4)

Milk yield, liters/day

9.22 ± 0.43

8.46 ± 0.41

Of the 82 calves which were reported to have been born alive ( January 2008 to December 31), 39 were from Hai and 43 from Meru giving an overall birth rate of 39%, which was comparatively lower than the figure estimated in East and Southern African regions( Swai et al 2005; French et al 2001). No animals were reported to have aborted during the same period. Births were reported to occur in all months of the year. A substantial proportion of calves were born during the period of April through July, however, there was no discernible pattern.

 

Estimation and factors influencing calving interval

 

The overall mean (mean ± SE) estimated CI of the cows that had calved more than once was 525±18 days. Mean calving interval was significantly higher in Meru (530±28) than in Hai (518±22) (P<0.05). The estimated mean calving interval was longer than that previously reported for cows kept in smallholder dairy units in Tanzania (Table 7).

Table 7: Reported initiation of ovarian activities(days) and calving intervals(days) in crossbred dairy cows in Tanzania

 

Production system

Type of animals

Initiation of ovarian activity and CI, days

(mean ± se)

Reference

1

Mixed smallholder rural/peri/urban dairy production  system-coastal north eastern Tanzania

Crossbred

122 ± 4.3

500 ±12.2

Swai et al 2005

2

Mixed smallholder rural  dairy production  system - eastern Usambara mountains Tanzania

Crossbred

108 ± 6.7

476 ± 14

Swai et al 2007b

3

Mixed smallholder  rural dairy production system - north western - Kagera Tanzania

Crossbred

197 ± 1.8

480 ± 2.4

Asimwe & Kifaro 2007

4

Mixed smallholder peri-urban dairy production  system- eastern Morogoro ,Tanzania

Crossbred

152 ± 6.1

477 ± 3.2

Nkya  et al 1999

5

Mixed smallholder rural dairy production  system - southern highlands  Mufindi, Iringa  Tanzania

Crossbred

(Ayr x Boran)

101 ± 3.6

402.6 ± 3.0

Chenyambuga & Mseleko  2009

6

Mixed smallholder peri/urban dairy production  system- eastern Dar-es-Salaam ,Tanzania

Crossbred

120 ± 2.8

450 ± 7.4

Kivaria et al 2006

7

Mixed smallholder peri/urban dairy production  system- north eastern Tanga ,Tanzania

Crossbred

276 ± 6.8

562 ± 13.2

Msangi et al 2005

Such a long calving interval and associated low calf crop, indicates poor reproduction performance resulting in lost farm income since cows spend a greater portion of their lactation at low production levels. The variables found to be significantly associated with variation in long calving interval in the most parsimonious multivariable regression models are shown in Table 8. Of the several farm and animal-level variables (n = 47) investigated; only three were significantly associated with long CI in the multivariable logistic regression models. Animals having better body condition (>3 BSC) were three fold (OR =3.8, P = 0.048) more likely to have a reduced calving interval below the recommended 430 days under tropical condition (Mujuni 1991). Low body weights were associated with extended long calving intervals (β for weightcentre2 = 0.01, P = 0.044). Of the farm related variables; unavailability of bulls was strongly associated with LCI (OR = 2.8; P = 0.023). Un-availability of bulls as a variable was strongly confounded by other variables such as bull being far away, access to AI on time and high breeding cost. Only 18 % of the farms were far than 4 km from the bull source.

Table 8: Variables associated with LCI in dairy cattle in multivariable logistic regression models- adjusted for farm effects

Variable

β (SE)

Wald P

LRS

LRP

Odds Ratio (95% CI)

Constant

-0.16 (0.26)

 

 

 

 

[Condition score centred]2

1.33 (0.68)

0.048

3.6

0.046

3.8 (1.04-14.05)

[Body weight centred]2

0.01(0.005)

0.044

4.0

0.043

1.2(1.01 -1.54)

Unavailability of bulls

1.03(0.46)

 

 

0.023

2.8(1.13-7.02)

Random (farm) effect

0.00 (0.41)

 

 

 

 

β = Coefficient of regression, SE = Standard error of coefficient, OR = Odd ratio, CI = Confidence Interval of OR, P = level of significance, LRS = Likelihood ratio statistic, LRP =Likelihood ratio p value

Other factors (univariate analysis) which marginally explain the problem of LCI were poor heat detection (LRS =3.25; P = 0.07), high breeding cost (LRS = 4.23, P = 0.038) and breeding or inseminating a cow within 6 hours after the onset of heat signs (LRS = 3.63, P = 0.047).

 

The mean CI estimated in this study (525 days) for dairy cows was higher than 488 and 457 days respectively, reported for F1 Friesian and Zebu and 3/4 Friesian and 1/4 Zebu crossbred dairy cows of smallholder dairy farms in Malawi (Agyemang and Nkhonjera 1990; Banda et al 2012) and 498 days in Vihiga district Kenya (Ongadi et al 2007). This value was similar to the values reported by Haile-Mariam et al (1993) at Abernossa Ranch, Ethiopia. On the other hand, CI found in the present study is higher than 351 to 398 days reported by Negussie et al (1998). The differences in the reproductive performance of crossbred cows reported by the different researchers might be attributed to the existing differences in nutritional, breeds and reproductive management among the smallholder dairy producers in different parts of the tropics. Several researchers suggested that differences in management might have accounted for the observed differences on LCI (Swai et al 2005; Lobago et al 2006).


Conclusions and Recommendations


Acknowledgments

The author would like to thank farm owners for their cooperation for being interviewed and NZARDEF Project No LP/09/05 for financial support during the course of this work. District Veterinary Officers from the two districts are acknowledged for providing the list of dairy farmers and cooperation. This work is dedicated to late Ali Malima (co-author) who passed away in the course of preparing this article. This paper is published with the permission from the Director of veterinary services in Tanzania.


References

Agyemang K and Nkhonjera L P 1990 Productivity of Crossbred Cattle on Smallholder Farms in Southern Malawi. Tropical Animal Health and Production 22: 9-16 

Asimwe L and Kifaro G C 2007: Effect of breed, season, year and parity on reproductive performance of dairy cattle under smallholder production system in Bukoba district, Tanzania. Livestock Research for Rural Development. Volume 19, Article #152. Retrieved August 9, 2013, from http://www.lrrd.org/lrrd19/10/asim19152.htm 

Banda L J, Kamwanja L A, Chagunda M G G, Ashworth C J and Roberts D J 2012 Status of dairy cow management and fertility in smallholder farms in Malawi. Tropical Animal Health and Production 44(4):715-27 

Chenyambuga S W and Mseleko K F 2009: Reproductive and lactation performances of Ayrshire and Boran crossbred cattle kept in smallholder farms in Mufindi district, Tanzania. Livestock Research for Rural Development. Volume 21, Article #100. Retrieved January 9, 2013, from http://www.lrrd.org/lrrd21/7/chen21100.htm 

Cytel Software Corporation 1999 Statistics and Epidemiology Research Corporation, version 2.0. Seattle, USA

Doto S P, Kimambo A E, Mgheni D M, Mtenga L A, Laswai  G H, Kurwijila L R, Pereka A E, Kombe R A, Weisberg M R, Hvelplund T, Madsen  J and Petersen P H 2004  Tanzania Feedstuff Table for Ruminants, ENRECA (SUA-MU/RVAU-DIAS) and ASLIP Projects, pp 12-16

Epi-info 1996 Centre for disease control, version 6.04d, Atlanta, USA and Geneva, Switzerland

FAO 2006. Tanzania – Agricultural census 2002-2003 – Main results. Small Holdings Sector. In: World census of agriculture – main results by country. Available 8 December 2013 from: http://www.fao.org/fileadmin/templates/ess/documents/world_census_of_agriculture/main_results_by_country/Tanzania_2002-03F_2.pdf 

French N P, Tyrer J and Hirst W M 2001 Smallholder dairy farming in the Chikwakwa communal land, Zimbabwe: birth, death and demographic trends. Preventive Veterinary Medicine 48: 101-112

Haile-mariam M, Banjaw K, Gebre-Meskel J and Ketema T 1993 productivity of Boran Cattle and Their Friesian Crosses at Abernossa Ranch, Rift Valley of Ethiopia. I. Reproductive Performance and Pre-weaning Mortality. Tropical Animal Health and Production 25: 239-248 

Kivaria F M, Noordhuizenm J P T M and Msami H M 2006 Risk factors associated with the incidence rate of clinical mastitis in smallholder dairy cows in Dar es Salaam region of Tanzania. The Veterinary Journal 173 (3): 623-629

Kristula M A, Curtis C R, Galligan D T and Bartholomew R W 1992 Use of repeated measures logistic regression model to predict chronic mastitis in dairy cows. Preventive Veterinary Medicine 14: 57-68.

Leslie J, Swai E S, Karimuribo E and Bell C 1999 Tanga and Southern Highland dairy development programmes: Socio-economic aspects and farmer perception of dairy cattle keeping and animal diseases, DFID/NRRD–Animal Health Research Programme, pp10-26

Lobago F, Bekana M, Gustaffson H and Kindahi H 2006 Reproductive performance of dairy cows in smallholder  proction system in Sellale, Central Ethiopia. Tropical Animal Health and Production 38: 333-342.

Maleko D D, Mbassa G N, Maanga W F and Sisya E S 2012 Impacts of Wildlife-Livestock Interactions in and around Arusha National Park, Tanzania. Current Research Journal of Biological Sciences 4(4): 471-476

MOAC/SUA/ILRI 1998 The Tanzanian Dairy Sub-Sector: A Rapid Appraisal: Volumes 1-3. Collaborative Research Reports of the Ministry of Agriculture and Co-operatives (Tanzania), Sokoine University of Agriculture (Tanzania) and the International Livestock Research Institute. Nairobi (Kenya)

Mlambiti M E, Edelsten P and Coyler D 1982 Economic analysis of the traditional farming systems of the Kilimanjaro Region Tanzania. IAF Publication No 85. University of West Virginia, Morgantown, West Virginia, USA

Msangi B S J, Bryant M J and Thorne P J 2005 Some factors affecting reproductive success in crossbred dairy cows on smallholder farms in coastal North-East Tanzania. Tropical Animal Health and Production 37 (5): 413- 426

Mtengeti EJ, Phiri E C J H, Urio N A, Mhando D G, Mvena Z, Ryoba R, Mdegela R, Singh  B R, Mo M, Wetlesen A, Lorken 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

Mujuni F P 1991 Postpartum Ovarian Disorders Affecting Dairy Cows in Selected Areas of Tanzania. MSc Thesis, Sokoine University of Agriculture, Tanzania pp 128

Munster B V 1997 A cow is a sign of Development, Report of the TDDP social economic survey, Tanga, Tanzania.

Nicholson M J and Butterworth M H 1986 A guide to condition scoring of zebu cattle, (ILCA Monograph; ILCA Addis Ababa, Ethiopia) 

Nkya R, Aboud A A and Kessy B M 1999 Improving productivity of smallholder dairy cattle in peri urban Morogoro Tanzania. In: Proceedings of the final Research Co-ordination Meeting of a Co-ordinated Research Project FAO/IAEA Vienna, Austria, 7-11 September 1998 pp. 79-86

Nkya R, Kessy B M, Lyimo Z C, Msangi B S J, Turuka F and Mtenga K 2007 Constraints on smallholder market oriented dairy systems in the north eastern coastal region of Tanzania. Tropical Animal Health and Production 39(8): 6 27-636 

Negussie E, Brannang E, Banjaw K and Rottmann O U 1998 Reproductive performance of dairy cattle at Assella livestock farm. Arsi. Ethiopia. I: Indigenous cows versus their F1 crosses. Journal of Animal Breeding and Genetics 115: 267-280 

Ongadi P M, Wakhungu J W, Wahome R G and Okitoi L O 2007: Characterization of grade dairy cattle owning households in mixed small scale farming systems of Vihiga, Kenya. Livestock Research for Rural Development. Volume 19, Article #43. Retrieved October 9, 2013, from http://www.lrrd.org/lrrd19/3/onga19043.htm 

Swai E S and Karimuribo E D 2011 Smallholder dairy farming in Tanzania: current profiles and prospects for developments. Outlook on Agriculture 40(1): 21–27 

Swai E S, Bryant M J, Karimuribo E D, French N P, Ogden N H, Fitzpatrick J L and Kambarage D M 2005 A cross-sectional study of reproductive performance of smallholder dairy cows in coastal Tanzania. Tropical Animal Health and Production 37: 513–525

Swai E S, Kyakaisho P and Ole Kawanara M S 2007a Methods of body condition scoring system: an important management decision support tool for resource-poor smallholders dairy farmers in Tanzania. Tanzania Veterinary Journal 24(2): 112-121

Swai E S, Kyakaisho P and Ole-Kawanara M S 2007b: Studies on the reproductive performance of crossbred dairy cows raised on smallholder farms in eastern Usambara mountains, Tanzania. Livestock Research for Rural Development. Volume 19, Article #61. Retrieved August 9, 2013, from http://www.lrrd.org/lrrd19/5/swai19061.htm 

URT, United Republic of Tanzania 2002 Population and household census. Available at http: //  www.tanzania.go.tz/cencus/table5.htm.2002; Accessed December, 2013 

URT, United Republic of Tanzania 2012 National Sample Census of Agriculture 2007/2008 Small Holder Agriculture Volume III: Livestock Sector – National Report (National Bureau of Statistics, Dar es Salaam, Tanzania), pp 8-10 

Van Niekerk A and Louw B P 1990 Condition scoring of beef cattle, Pretoria, Department of Agriculture Development, South Africa, pp 4-8


Received 19 January 2014; Accepted 11 March 2014; Published 1 June 2014

Go to top