Smallholder dairy production in high altitude Nyandarua milk-shed in Kenya: Status, challenges and opportunities

J M K Muia, J N Kariuki, P N Mbugua, C K Gachuiri, L B Lukibisi, W O Ayako and W V Ngunjiri**

KARI-Naivasha Research Centre, P.O. Box 25 - 20117 Naivasha, Kenya
* Department of Animal Production, University of Nairobi, P.O. Box 19832 - 00100 Nairobi, Kenya.
** Ministry of Livestock Development, P.O. Box 476 - 20300 Nyandarua, Kenya
john.muia@yahoo.com

Abstract

A stratified sampling method was used to select 156 dairying households from representative Divisions in Nyandarua County. The stratification was based on cattle grazing systems (CGS) and agro-ecological zones (AEZs) across the Divisions. The objectives of the study were to assess status of smallholder dairy cattle production in relationship to CGS and AEZ, major challenges facing smallholder dairy production, and the opportunities for improvement. Data collected included the characteristics of the farm, family, farmer, feeds and feeding, dairy cattle and their performance, milk uses and markets, and the dairy production services. The information on the challenges facing dairy production and the opportunities for improvement was obtained from discussions with livestock extension workers, dairy co-operatives, milk processors, and from secondary sources.

The present results indicated that the average farm size was 3.5 Ha and 41, 38, and 44% of the households fed dairy stock with improved fodders, grass hay, and concentrate supplements, respectively. Among the households, about 44, 38 and 32% had access to artificial insemination (AI), extension, and all weather roads services, respectively. Households keeping crosses of the dairy breeds were 59% while the average herd size was 5.3 heads consisting of 40% cows in milk. The average calf live-weight gain was 322g/ day and milk yield per cow was 8.4kg/day. About 65% of the milk was marketed at an average price of 15.00 KES/kg, equivalent to 0.205 US$/kg. As the levels of dairy intensification increased, there were significant increase in milk production per hectare and decrease in calf live-weight gains (P<0.05). On the other hand, as the level of agricultural potential increased, there were significant decreases in milk production and marketed milk per farm (P<0.05).

It was concluded that smallholder dairy cattle production was below the potential for Nyandarua County and was influenced by the CGS and AEZs. The major challenges in smallholder dairy production included poor road network and milk marketing, high costs and inaccessibility of dairy production inputs and support services, inappropriate dairy production technologies, and limited value addition of milk.

Key words: Agro-ecological zones, cattle grazing systems, household characteristics

Introduction

Kenya’s dairy industry, the single largest livestock production sub-sector contributes 14% of the agricultural gross domestic product (GDP) and 3.5% of the total GDP (Muriuki et al 2003). The industry plays an important role in food security, employment creation, income generation, and enhances the livelihoods of dairy farmers, traders, processors and all participants engaged in the entire milk supply chain. The total dairy herd estimated at 3.4 million heads produces about 3.1 billion litres of milk annually (Kenya National Bureau of Statistics (KNBS) 2010; Ministry of Livestock and Fisheries Development (MoL&FD) 2003). Dairy production is dominated by smallholders who own about 98% of the total dairy herd (Peeler and Omore 1997). Smallholder dairying households estimated to number over 1.5 million households, account for more than 85% of the annual total milk production and 80% of the 1.8 billion litres of milk marketed annually (MoL&FD 2003; Stall et al 2001). Over the years, significant changes in the traditional dairying have occurred resulting in a major shift towards market-oriented smallholder production. This has been possible mainly due to the suitable climatic conditions, significantly improved fodder technology and dairy cattle population, high urban population and incomes and the high consumption of milk and dairy products. In addition to the economic importance of milk, cattle manure is used to improve soil fertility resulting in increased pasture/fodder production on smallholder farms.

The country is generally self-sufficient in milk and dairy products. However, the demand for milk and dairy products in developing countries is estimated to increase by 25% by 2025 (Delgado et al 1999), mainly due to human population growth, further urbanization, increased disposable income, greater diversity of food products to meet nutritional needs, and increased opportunities for domestic and external trade. Indeed, dairy imports in developing countries may reach 38.9 billion litres of milk equivalent by 2030 (Food and Agriculture Organization (FAO) and International Dairy Federation (IDF) 2004). Fortunately, the country has the potential to increase milk production from the current 4.2 billion litres in 2009 to over 5.0 billion litres in 2014 (Cherono 2005). Milk production and market opportunities represent exciting challenges for smallholders in the country and if these potential productions and markets have to be exploited, it will require expansion of specialized dairy cattle population, intensification in terms of inputs, value addition of milk and dairy products, and good market linkages for milk sales and input acquisition.

Nyandarua County produces the highest amount of milk due to its higher population of dairy cows as compared to the other regions in Central Kenya (MoL&FD 2007). However, reports for Central Kenya indicates that dairy production potential for Nyandarua County is the least exploited (Romney et al 2004; Staal et al 2001; Schreiber 2000; Baltenweck et al 1998). There is therefore a need to improve the efficiency of dairy production and marketing for equitable distribution of income and hence poverty alleviation among households especially in the rural areas in line with the Vision 2030 (Government of Kenya (GOK) 2007). Recommendations for dairy development must therefore be based on the prevailing dairy farming circumstances, opportunities and challenges in the region. Detailed information on smallholder dairying for the region and within the cattle grazing systems (CGS) and the agro-ecological zones (AEZs) is therefore needed.

The objectives of this study were therefore to assess status of smallholder dairy cattle production in relationship to CGS and AEZ, challenges, and the opportunities for improvement in Nyandarua County.

Materials and methods

Study sites

Nyandarua County borders Laikipia County to the North, Nyeri and Muranga Counties to the East, Thika and Kiambu Counties to the South, and Nakuru County to the West. It covers an area of 3,304 km² and has a population density of 145 per km² and 104,401 households according to a report by the Central Bureau of Statistics (CBS) (2001). The study was conducted in five out of six administrative Divisions (sites) of the County (Ndaragwa, Ol-Joro-Orok, Ol-Kalou, Kipipiri and North Kinangop). The soils in the upper highlands are imperfectly to well drained, deep to very deep, dark brown to very dark grayish brown, firm clay and silt loam clay. However, in the lower highlands, the soils are well drained, moderately to very deep, dark grayish brown to dark reddish brown, clay loam to clay with humic top soil. Details of the soils and agro-climatic characteristics of the different AEZs are shown in Table 1. Among the different dairy cattle production systems, three distinct cattle grazing systems associated with dairy intensification levels were identified in Central Kenya (Romney et al 2004; Staal et al 2001; Schreiber 2000; Baltenweck et al 1998).

Household selection and sampling procedures

The dairying households were selected from five representative sites. The 85,739 total households for the selected sites were obtained from the 1999 population and housing census in Kenya (CBS 2001). Initial selection of households was purposive based on those keeping pure dairy cattle and their crosses, and willing to participate in the study. Since the percentage of dairying households in Central Kenya were estimated at 56% (Omore et al 1999), the estimated dairying households for the selected sites were 48,013. A simple stratified sampling method was used to select the dairying households across the sites. Stratification was done by grouping dairying households within sites into relatively homogeneous sub-groups (stratum) before sampling. The distinct three CGS were zero grazing, semi-zero grazing and free grazing while the four AEZs were AEZ I, AEZ II, AEZ II and AEZ IV which formed the stratum. For effective stratification, minimum variability within strata and maximum variability between strata were maintained.

The AEZs were based on the “Exploratory Soil Map and Agro-climatic Zone Map of Kenya” by Sombroek et al (1982). However, the classification of households into different CGS was based on dairy intensification levels; intensive stall feeding of cattle (zero grazing), semi-intensive feeding of cattle where field grazing was done during part of the day while stall feeding was done during the remaining part of the day (semi-zero grazing), and the extensive feeding of cattle where field grazing was done all the day (free grazing). Selection of household within a stratum was by simple random sampling.

The sample size was estimated using the following formula by Israel (2009):

N = Z²p(1-p)

E²

Where, N = Sample size required.

Z = Confidence level at 95% (standard value of 1.96).

p = Estimated proportion of attribute that is present in the population

E = Desired level of precision to detect difference among strata observations.

The sample size was estimated by assuming variability in the proportion of dependent variables (p) of 0.5 and a desired level of precision of 15% which was potentially needed to estimate a difference among the strata. The obtained sample size was therefore multiplied by 3 to correct for design effect and accommodate a comparative analysis of the strata, and increased by 20% to adjust for non-response or recording errors. The sample size was then rounded up to the closest number that matches well with the number of strata (3 CGS and 4 AEZs) to be surveyed. A total of 156 households were therefore required for the survey. Sample size per stratum was estimated by dividing total households required by number of strata while sample size per site was taken as the proportion of dairying households as estimated from the 1999 population and housing census in Kenya (CBS 2001).

Data collection and analysis

A structured questionnaire was designed to collect quantitative and qualitative data. The questionnaire was pre-tested on-farm and the necessary amendments were incorporated before wide scale household direct interviews were conducted. Each questionnaire captured household data such as site, CGS, AEZ, farm, family, farmer, dairy cattle, feeds and feeding, cattle performance, various uses of milk, marketing channels, and dairy cattle production services (Tables 2 to 7). For ease of comparison, each herd size was converted into tropical livestock units (TLU) according to Anon (1992). Secondary information on the challenges facing smallholder dairy cattle production and the opportunities for improvement was obtained from livestock extension workers, management staff of farmer dairy co-operatives and milk processors, and from secondary sources of data.

Out of 156 households selected for the study according to Israel (2009), 138 (0.3% of the dairying households) returned filled questionnaires. Data from the 138 questionnaires were therefore analyzed using the General Linear Model of Statistical Analysis Systems (SAS) (1999). The means of the dairy cattle production and marketing parameter for the 3 CGS and the 4 AEZs were separated using Least Significance Difference at 95% confidence level.

Table 1 Soils and agro-climatic characteristics of the different agro-ecological zones in Nyandarua County
Variable	AEZ I	AEZ II	AEZ III	AEZ IV
General description	Afro-Alpine highlands and upper highlands	Upper highlands	Lower highlands	Lower highlands
Average number of growing days	365	290 - 365	235 - 290	180 - 235
Soils	Dystic Histosols and verto-luvic Phaeozems	verto-luvic and chromo-luvic Phaeozems	Chromo-luvic Phaeozems	chromo-luvic and ando-luvic Phaeozems
Climatic designation	Humid (cold to very cold)	Sub-humid (very cool)	Semi-humid (cool)	Semi-humid to semi-arid (fairly cool)
Mean temperature (⁰C)	10 - 12	12 - 14	14 -16	16 -18
Mean maximum temperatures (⁰C)	16 - 18	18 - 20	20 - 22	22 - 24
Mean minimum temperatures (⁰C)	4 - 6	6 - 8	8 -10	10 -12
Night frosts	Very common	Common	Rare	Very rare
Annual average rainfall (mm)	1,100 - 2,700	1,000-1,600	800 -1,400	600 -1,100
Rainfall/evaporative demand (%)	> 80	65 - 80	50 - 65	40 - 50
Cattle carrying capacity (LU/Acre)	> 8	4 - 8	2 - 4	1 - 2
Divisions	Kinangop	Kipipiri and Ol-Joro-Orok	Ndaragwa, Ol-Joro-Orok, OlKalou and Kipipiri	Ndaragwa and Ol-Kalou
% of the County	40	25	25	10
Adapted from Sombroek et al 1982; AEZ I – agro-ecological zone one; AEZ II – agro-ecological zone two; AEZ III – agro-ecological zone three; AEZ IV – agro-ecological zone four; LU-Livestock Units

Results and discussion

Characteristics of smallholder dairy cattle production in Nyandarua County

The characteristics of smallholder dairy cattle production and marketing in Nyandarua County are shown in Tables 2 to 4.

Table 2 Characteristics of the households and fodder production for the different cattle grazing systems
Variable	ZG	SZG	FG	SED	Mean± SD
No. of households	45	47	46
Size of farm, Ha	2.47^b	2.25^b	5.82^a	0.651	3.53 ± 3.09
Farm under natural pastures, %	23.8^a	23^a	22.5^a	1.8	23.1 ± 8.64
Farm under improved fodders, %	7.0^a	6.85^a	6.63^a	1.84	6.83 ± 8.8
HH with improved fodders, %	44.4^a	36.2^a	41.3^a	10.1	40.6 ± 48.2
HH with grass hay, %	37.8^a	38.3^a	39.1^a	10.0	38.4 ± 48.1
HH with silages, %	28.9^a	12.8^b	8.7^b	7.72	16.7 ± 37.0
HH supplementing concentrates, %	51.1^a	36.2^a	45.7^a	10.1	44.2 ± 48.6
HH access to extension services, %	53.3^a	29.8^b	30.4^b	9.98	37.7 ± 47.8
HH access to all weather roads, %	33.3^a	25.5^a	37.0^a	9.43	31.9 ± 45.2
HH access to AI services, %	46.7^a	40.4^a	43.5^a	10.1	43.5 ± 48.5
ZG – zero grazing; SZG – semi-zero grazing; FG – free grazing; SED – standard error of difference; SD – standard deviation; HH=Households; A.I. – artificial insemination; means in a row and bearing different superscripts are different at P<0.05

In this study, average farm size of 3.5 Ha was lower than the range of 4.5 - 8 Ha previously reported for this area by Romney et al (2004), Schreiber (2000) and Baltenweck et al (1998). This implied that the size of farms had decreased over time possibly due to increased human population.

Table 3 Characteristics of the household families for the different dairy cattle grazing systems
Variable	ZG	SZG	FG	SED	Mean± SD
No. of households	45	47	46
Family size	5.91^ab	6.57^a	5.67^b	0.512	6.06 ± 2.46
Males in a family, %	50.3^a	49.2^a	47.5^a	3.76	48.9 ± 18.1
Male HHH, %	51.1^a	48.9^a	47.8^a	10.6	49.3 ± 50.7
Age of HHH, years	43.2^a	43.6^a	43.5^a	1.67	43.4 ± 7.97
HHH of above primary education, %	24.4^a	25.5^a	26.1^a	9.01	25.4 ± 43.2
HHH of primary education, %	37.8^a	38.3^a	47.8^a	10.2	41.3 ± 49.1
HHH of below primary education, %	37.8^a	36.2^a	26.1^a	9.98	33.3 ± 47.8
ZG – zero grazing; SZG – semi-zero grazing; FG – free grazing; SED – standard error of difference; SD – standard deviation; HHH – household heads; means in a row and bearing different superscripts are different at P<0.05

Forty-one percent of the households had improved fodders, 38% fed cows with grass hay, 38 % accessed extension services, 44% supplemented cows with concentrates and 44% accessed artificial insemination (AI) services. However, only 7% of the farm size had improved fodders, and 17% and 32% of the households fed silages to cows and accessed all weather roads, respectively. The 6.1 heads per family (Table 3) was in agreement with the CBS in Kenya (2001) but slightly lower than the values previously reported for the area (Romney et al 2004; Makenzi 2008). The male to female (49:51) ratio for the family and the household heads were comparable to other reports for the area (CBS 2001; Makenzi 2008). The average age of household heads of 43.4 years was comparable to the average age reported by Makenzi (2008). Sixty-five percent of the household heads had primary education and over, whereas illiteracy rate was 35%.

Table 4 Characteristics of milk production and marketing for the different dairy cattle grazing systems
Variable	ZG	SZG	FG	SED	Mean± SD
No. of households	45	47	46
Pure dairy breeds, %	37.8^ab	51.1^a	32.6^b	10.4	40.6 ± 49.8
Size of herd	5.26^a	5.38^a	5.24^a	0.751	5.29 ± 3.54
Cows in milk	2.13^a	2.38^a	2.4^a	0.259	2.31 ± 1.22
Stocking rate, TLU/Ha	4.18^a	3.74^a	1.43^b	0.726	3.08 ± 3.39
Rate of concentrates, kg/cow/day	3.56^a	2.78^b	2.95^b	0.306	3.09 ± 1.26
Calf weaning age, months	4.23^a	4.17^ab	3.9^b	0.197	4.1 ± 0.879
Calf live-weight gain, g/day	293^b	313^ab	362^a	40.3	322 ± 180
Milk production, kg/farm/day	17.5^a	18.1^a	19.0^a	2.14	18.2 ± 10.1
Milk production, kg/Ha	14.8^a	14.4^a	4.89^b	3.42	11.3 ± 16.0
Milk yield, kg/cow/day	8.43^a	8.21^a	8.63^a	0.698	8.42 ± 3.29
Marketed milk, kg/farm/day	11.0^a	12.2^a	12.7^a	1.66	12.0 ± 7.73
Price of milk, KES/kg*	15.1^a	14.9^a	15.1^a	0.413	15.0 ± 1.87
HH access to formal milk markets, %	35.6^a	38.3^a	34.8^a	10.3	36.2 ± 49.3
ZG – zero grazing; SZG – semi-zero grazing; FG – free grazing; SED – standard error of difference; SD – standard deviation; TLU – total livestock units; KES – Kenya shilling; HH – households; means in a row and bearing different superscripts are different at P < 0.05; *1 US$ = 73 KES

Data in this study indicated that the crosses between the main dairy breeds (Friesians and Aryshires) were 59% while the rest of the dairy cattle were pure dairy breeds (Table 4). The herd size of 5.3 heads was higher than the value previously reported for this area (Schreiber 2000). This implied that the size of the herd had increased over time and because of the decreased land size, there was even more pressure on the scarce feed resources than was previously the case. The cows in milk represented 40% of the herd while stocking rate was 3.1 TLU/Ha. In this study, the concentrates were supplemented at daily rate of 3.1 kg/cow. The weaning age of 4.1 months was higher than the recommended 3 months and the daily live-weight gain of 322 g was lower than the expected range of 400-500 g for dairy calves under good management (Kiragu et al 2008). Milk production per farm of 18.2 kg (Table 4) was higher than the 10.0 kg reported for smallholder farms in Kenya (Thorpe et al 2000) while the milk production per hectare of 11.3 kg was higher than values previously reported for the area (Romney et al 2004). The daily milk yield of 8.4 kg/cow was comparable to the yields attributable to under-nutrition of dairy cattle in smallholder farms in Kenya (Omore et al 1999).

About 65% of the milk production in this study was marketed (Table 4) which was an improvement from 50% previously reported for the area by Romney et al (2004).This would be attributable to the increased marketing of milk through the informal sector. However, as expected the marketed milk was lower than the 90 % reported for Kiambu County which is relatively closer to Nairobi (Staal et al 1998). The milk price of 15.00 KES/kg (0.205 US$) was very low and would be comparable with the country’s average in the last nine years (Thorpe et al 2000). The low milk price was a reflection of the high costs of transportation due to poor road infrastructure and long distance to markets. In contrast, milk prices in Kiambu County of 18.00-20.00 KES/kg (Mburu et al 2007) were higher due to the comparatively better road infrastructure and its proximity to Nairobi market. However, the milk price in this study was an improvement over the 14.00 KES/kg reported previously for the area (Romney et al 2004; Staal et al 2003) due to the increased involvement of the informal milk marketing agents who purchased milk at relatively higher prices than through the formal sector.

The higher milk productions per farm and per hectare as compared to other areas in Kenya were mainly due to the higher number of cows in milk but not the milk yields per cow. However, the low growth rate of calves and milk yield per cow was mainly due to overstocking which resulted in inadequate intake of feeds by the dairy stock. Due to inadequate feeds, calves were weaned later than the recommended age of three months under good plane of nutrition. Also, due to poor access to AI services farmers relied heavily on natural service (bull). Natural services may be associated with inbreeding and sexually transmitted diseases and hence genetically inferior animals. This may negatively affect performance of the dairy stock. Also, the low adoption by farmers of fodder production and conservation technologies, low rates of concentrate supplementation, low formal education of household heads, and the poor access to extension services may also have contributed to the low animal performance.

Smallholder dairy cattle production among cattle grazing systems

Table 2 shows the characteristics of the smallholder dairy farms, fodder production and conservation, and the access to dairy cattle production services among the grazing systems in Nyandarua County. In agreement with Romney et al (2004) the average farm size in the present study tended to decrease with increasing levels of dairy intensification. The farm size for the free grazing households was significantly higher than for the zero grazing and semi-zero grazing households (P<0.05). On the contrary, silage making and access to extension services for the zero grazing households was significantly higher (P<0.05) than for the semi-zero grazing and free grazing households. However, gender, ages and education levels of household heads (Table 3), fodder production, use of grass hay, concentrate supplementation, and access to all weather roads and AI services were not significantly different (P<0.05) among the grazing systems. In contrast to this study, fodder production increased with increasing levels of dairy intensification in other areas of the country (Mwendia et al 2007).

As shown in Table 4, the live-weight gains of calves tended to decrease significantly (P<0.05) with increasing levels of dairy intensification. However, in a different study in Kenya, live-weight gains of dairy calves did not differ significantly (P > 0.05) among the intensification levels (Lanyasunya et al 2001). The decline in calf live-weight gains would be attributed to the decline in average farm size, percentage of literate farmers and a corresponding increase in stocking rates as levels of intensification increased. Overstocking may have led to inadequate forage feeding to calves resulting in poor rumen development, inefficient utilization of solid feeds, low growth rates and hence late weaning (Kiragu et al 2008). For optimal management of calves, the farmer must have basic formal education with good understanding of modern young stock rearing techniques. Therefore, adoption of agricultural technologies and hence calf performance was expected to be higher among the well educated than the poorly educated household heads. Although the calf weaning ages for the zero grazing and the semi-zero grazing households were significantly higher than for the free grazing households, the live-weight gains for the zero grazing and the semi-zero grazing households were significantly lower than for the free grazing households (P<0.05). The percentages of pure dairy breeds were significantly higher for the zero grazing and the semi-zero grazing than for the free grazing households (P < 0.05). Likewise, the average stocking rates were significantly higher (P<0.05) for the zero grazing and the semi-zero grazing than for the free grazing households.

Generally, milk production per hectare tended to increase with increasing levels of dairy intensification (Table 4) in agreement with a previous report (Romney et al 2004). This would mainly be due to the increase in access to extension services as levels of dairy intensification increased. Due to increased access to extension services, rates of concentrate supplementation, and use of silages tended to increase. The milk production per hectare was significantly higher for the zero grazing and the semi-zero grazing than for the free grazing households (P<0.05). On the other hand, the rate of concentrate supplementation was significantly higher for the zero grazing than the semi-zero grazing and free grazing households (P<0.05). The herd size, number of cows in milk, milk production/farm, milk yield/cow, marketed milk, milk prices, and access to milk marketing channels were not significantly different among the grazing systems (P>0.05). However, in previous studies, the size of the herd and cows in milk decreased (Mwendia et al 2007) while the milk production per cow increased (Lanyasunya et al 2006) with increasing levels of dairy intensification mainly due to improved nutrition.

Smallholder dairy cattle production among agro-ecological zones

Table 5 shows characteristics of the farms, fodder production and conservation, and access to dairy cattle production services among the AEZs in Nyandarua County. In Kenya, the agricultural potential of land generally decreases with increasing AEZs (Sombroek et al 1982). As expected, farm sizes tended to decreased with increasing agricultural potential of the land. The farm size for AEZs IV and III were significantly higher than for AEZ I households (P<0.05). However, the farm size under improved fodders, households with improved fodders, use of grass hay, household supplementing concentrates, access to all weather roads, and access to AI services tended to decrease as the agricultural potential of land increased. Since the study was conducted in water catchments, agricultural activities tended to decrease from AEZ IV to AEZ I to minimize land degradation and to conserve the fragile environment through forest establishment (Kimigo et al 2008).

The farm sizes with improved fodders for AEZs III and II were significantly higher than for AEZ 1 households while improved fodders were significantly higher for AEZs IV, III and II than for AEZ I households (P<0.05). The use of grass hay for AEZs III and II were significantly higher than for AEZ I households (P<0.05). The use of concentrates was significantly higher for AEZ II than for AEZs III and I households (P<0.05). Access to all weather roads was significantly higher for AEZ IV than for AEZ II households while access to AI services was significantly higher for AEZ II than for AEZ I households (P<0.05). However, the percentage of farm under natural pastures, silage making, and the access to extension services were not significantly different among the AEZs (P>0.05).

Table 5 Characteristics of the households and fodder production for the different agro-ecological zones
Variable	AEZ I	AEZ II	AEZ III	AEZ IV	SED
No. of households	34	35	35	34
Size of farm, Ha	2.55^b	3.59^ab	4.08^a	3.89^a	0.752
Farm under natural pastures, %	22.2^a	23.1^a	23.4^a	23.5^a	2.08
Farm under improved fodders, %	3.24^b	9.2^a	8.43^a	6.32^ab	2.12
HH with improved fodders, %	17.7^b	48.6^a	57.1^a	38.2^a	11.6
HH with grass hay, %	20.6^b	48.6^a	51.4^a	32.4^ab	11.6
HH with silages, %	8.82^a	20.0^a	17.1^a	20.6^a	8.92
HH supplementing concentrates, %	29.4^b	62.9^a	37.1^b	47.1^ab	11.7
HH access to extension services, %	32.4^a	37.1^a	42.9^a	38.2^a	26.4
HH access to all weather roads, %	29.4^ab	20.0^b	37.1^ab	41.2^a	10.9
HH access to AI services, %	32.4^b	57.1^a	37.1^b	47.1^ab	11.7
AEZ I – agro-ecological zone one; AEZ II – agro-ecological zone two; AEZ III – agro-ecological zone three; AEZ IV – agro-ecological zone four; SED – standard error of difference; HH – households; A.I – artificial insemination; means in a row and bearing different superscripts are different at P<0.05

As shown in table 6, the education level of farmers tended to decrease as the agricultural potential of the land increased. The increased poor state of the roads as the agricultural potential increased was likely to have a negative effect on education levels of the household heads. The family members for AEZs II and III were significantly higher than for AEZ IV households (P<0.05). The household heads who attended school beyond primary were significantly higher for AEZ IV than for AEZs II and I (P<0.05) while the household heads with primary education were significantly higher for AEZ II than for AEZs I (P<0.05). However, the education below primary, age and sex ratio of household heads were not significantly different (P>0.05) among the AEZs.

Table 6 Characteristics of the household families for the different agro-ecological zones
Variable	AEZ I	AEZ II	AEZ III	AEZ IV	SED
No. of households	34	35	35	34
Family size	6.01^ab	6.49^a	6.46^a	5.21^b	0.591
Males in a family, %	48.1^a	46.9^a	54.0^a	46.8^a	4.35
Male HHH, %	41.2^a	42.9^a	57.1^a	55.9^a	12.2
Age of HHH, years	43.3^a	42.3^a	43.9^a	44.1^a	1.93
HHH of above primary education, %	20.6^b	17.1^b	25.7^ab	38.2^a	10.4
HHH of primary education, %	35.3^b	57.1^a	37.1^b	35.3^b	11.8
HHH of below primary education, %	44.1^a	25.7^a	37.1^a	26.5^a	11.5
AEZ I – agro-ecological zone one; AEZ II – agro-ecological zone two; AEZ III – agro-ecological zone three; AEZ IV – agro-ecological zone four; SED – standard error of difference; HHH – household heads; means in a row and bearing different superscripts are different at P<0.05

Table 7 Characteristics of milk production and marketing for the different agro-ecological zones
Variable	AEZ I	AEZ II	AEZ III	AEZ IV	SED
No. of households	34	35	35	34
Pure dairy breeds, %	41.2^a	34.3^a	42.9^a	44.1^a	12.0
Size of herd	4.56^b	4.74^ab	6.06^a	5.76^ab	0.868
Cows in milk	2.21^ab	1.97^b	2.43^ab	2.61^a	0.299
Stocking rate, TLU/Ha	3.08^a	3.57^a	2.72^a	3.03^a	0.839
Rate of concentrates, kg/cow/day	3.09^a	3.19^a	3.12^a	2.96^a	0.353
Calf age at weaning, months	4.17^a	4.1^a	4.2^a	3.93^a	0.227
Calf live-weight gain, g/day	289^a	343^a	323^a	334^a	46.6
Milk production, kg/farm/day	15.9^bc	15.6^c	19.7^ab	21.6^a	2.47
Milk production, kg/Ha	10.9^a	11.1^a	8.37^a	14.8^a	3.96
Milk yield, kg/cow/day	7.86^a	8.27^a	9.05^a	8.49^a	0.807
Marketed milk, kg/farm/day	10.4^b	10.2^b	13.0^ab	14.2^a	1.91
Price of milk, KES/kg*	14.9^a	15.0^a	15.4^a	14.8^a	0.477
HH access to formal milk markets, %	29.4^a	45.7^a	34.3^a	35.3^a	11.9
AEZ I – agro-ecological zone one; AEZ II – agro-ecological zone two; AEZ III – agro-ecological zone three; AEZ IV – agro-ecological zone four; SED – standard error of difference; TLU – total livestock units; KES – Kenya shillings; HH – Households; means in a row and bearing different superscripts are different at P<0.05; *1 US$ = 73 KES

The characteristics of the herd, breeds, stocking rates, performance of calves and cows, milk marketing channels and milk prices among the AEZs are shown in Table 7. As the agricultural potential of the land increased, the size of herd, cows in milk, milk production per farm and the marketed milk tended to decrease due to the decreasing farm sizes and the general use of land for agricultural purposes. The herd size was significantly higher for AEZ III than for AEZ I households while the number of cows in milk were significantly higher for AEZ IV than for AEZ II households (P<0.05). The milk production per farm and marketed milk were significantly higher for AEZ IV than for AEZ II and I households (P<0.05). However, households with pure dairy breeds, stocking rates, rates of concentrate supplementation, calf age at weaning, calf live-weight gains, milk production per hectare, milk yield per cow, price of milk and access to formal milk marketing channels were not significantly different among the AEZs (P>0.05).

In Kiambu County, milk yield per cow tended to increase as the agricultural potential of land increased (Mburu et al 2007). This trend was not evident in the present study because the milk yield per cow, milk production per hectare and the live-weight gains of calves were relatively lower for the zone expected to have the highest agricultural potential as compared to the other zones. The AEZ I was mainly found in areas of the County with poor adaptability of the common improved fodders and pastures due to cold weather and frequent frost bites. Because of the fragile nature of the land in AEZ I, protection of the environment and conservation of the biodiversity were achieved through controlled agricultural activities, forest establishment and water catchments. From this study, it was evident that dairy cattle production activities were relatively low in AEZ I as compared to the other zones.

Challenges in smallholder dairy cattle production

Poor road infrastructure

The poor state of the roads was evident from this study since only 30% of the households had access to good roads and hence could purchase inputs and market their farm produce throughout the year. During the rain seasons, most of the roads were impassable particularly in the upper highlands with firm clay and clay loam soils hence farmers were unable to sell their farm produce. Due to the poor road network and long distance to markets, cost of transportation was high rendering smallholder dairy production uncompetitive.

Poor marketing

Most of the milk produced during the wet season was not marketed due to the poor road network and long distance to the markets. Since milk is highly perishable and farmers did not have the means to invest in milk cooling equipments, the high volumes of milk produced during the wet season were therefore associated with high-post harvest losses. Only about 35% total milk production was marketed through the formal sector which is considered by farmers to be more reliable in terms of milk prices and payments for milk delivered than the informal sector. This was mainly due to low milk processing capacity of the formal sector. As a result, the only alternative was for farmers to sell the surplus milk through the informal sector at lower prices. In addition, poor organization of milk collection, processing and marketing systems seriously undermined the potential of smallholder dairy producers to exploit urban markets.

High costs and unavailability of inputs and support services

The increased costs of transportation and distribution systems due to the poor road network and long distance to markets resulted in high costs of inputs (supplements, animal drugs and vaccines, pesticides, fertilizers, and herbicides) and their unavailability. In addition, the high costs of other services such as AI, animal health, electricity supply, extension and training, and credit had a negative impact on dairy development in the study area. High cost and unavailability of electricity in rural areas reduced investments especially in cold storage facilities and processing of the highly perishable goods such as milk and dairy products. The cost of credit, limited use of land as collateral for financing farming, and the limited number of banks in the rural areas are some of the factors that made it difficult for farmers to access credit from formal banking industry.

Inadequate use of appropriate technologies

The high cost and inaccessibility of AI services caused about 60% of the households to use natural breeding methods and hence were unable to sustain genetic improvement. Natural breeding method resulted in genetically inferior animals due to inbreeding and the use of bulls of inferior genetic potential negatively affected performance of the offspring. On the other hand, since improved fodder production and conservation were low, the dairy stock relied mainly on inadequate and poor quality natural pastures with low levels of supplementation. The poor adaptability of common fodders and grasses due to low temperatures and frequent frost in upper highlands and frequent drought in lower highlands resulted in shortage of animal feeds and hence the farms were overstocked. The poor access to extension services, and the limited knowledge and skills on animal husbandry among the household heads due to the high levels of illiteracy (35%) resulted in poor performance of the dairy stock. Dairying was not competitive due to high costs of production and the use of inappropriate technologies, and hence poor performance of the sector.

Limited value addition of milk and dairy products

Most of the milk from this area was marketed fresh through the informal sector (65%). Since fresh raw milk is highly perishable, milk losses along the informal value chain were high resulting from spillage and spoilage due to the poor road network, long distance to markets, inadequate refrigeration, and lack of milk collection due to glut in the wet season. Also, due to inadequate regulations, poor hygiene of milk at all levels of production and marketing was a common problem. Consumption of fresh milk was therefore associated with health risks since it is an excellent media for bacteria and has the potential to transfer zoonotic diseases to consumers. Failure to meet international food-safety and quality standards due to the domination of milk marketing by the informal sector hampered efforts to participate in regional and international markets resulting in low milk prices and hence sub-optimal dairy production in the study area. On the other hand, marketing of milk through the formal sector was limited due to the high costs of processing, value addition and increased shelf-live and packaging of milk and dairy products.

Encroachment of the forests and water catchments by agricultural activities

The migration of farmers from the other overpopulated areas of Central Kenya into the study area has increased socio-economic activities resulting in expansion of human settlements, cultivation of land for crop/animal production, and deforestation particularly in the forested steep areas and water catchments. Furthermore, due to the high costs of fertilizers and the fact that manure collection and it use was not possible under the common extensive grazing systems among the households, poor soil fertility resulted in low fodder yields. If this trend continues unabated, environmental degradation, water pollution, poor conservation of fauna and flora and other associated environmental problems will be encountered in the near future.

Opportunities to improve smallholder dairying

High demand for milk and dairy products

The study area produces the highest amount of milk due to larger population of dairy cattle than any other region in central Kenya. Due to the low consumer prices, fresh milk can be marketed among the populous poor urban dwellers and the milk deficit rural areas. On the other hand, the increases in disposable income, and changes in consumer preferences (tastes) among the urban dwellers has created a high domestic demand for high value food items such as milk and milk products creating market opportunities for indigenous production. In addition, due to the large regional markets which have arisen through regional integration (East African Community, Common Market for Eastern and Southern Africa, African Growth Opportunity Act, World Trade Organization, African Caribbean and Pacific) and the preferential treatment provided to products from member countries there is great potential to improve smallholder dairy production and marketing in this area. Full exploitation of the existing and emerging milk and dairy product markets will broaden trade and income base for the area and the country in general. To effectively exploit these opportunities, the main challenges will be to improve quality and safety, increase efficiency and competitiveness in production and marketing of milk and dairy products.

Expansion and improvement of road network

The existing road network needs to be improved and expanded to reduce cost of dairy production and hence increase marketed milk beyond the 65% level reported in the study area. A lot of emphasis in improvements and expansions of the road network should be directed towards upgrading of the feeder roads which link the farms to the milk collection centres. Upgrading of the feeder roads which are impassable during the rain season will significantly increase the collection and marketing of milk from farm. Fast transportation and marketing is important due to the perishable nature of milk and its products. Also, transportation of inputs and other dairy production support services would benefit from expanded and improved road network. The road network can be improved and expanded not only by the central government but also by local communities through innovative partnerships including those with the private sector.

Marketing developments

The informal sector, controlling about 65% of the milk marketed dealt mainly with raw milk which was commonly used to make tea, coffee or as food snack and therefore did not require any processing. The higher preference by consumers for raw milk as compared to processed milk, provides an opportunity for the informal sector and hence the smallholder dairy production system to be competitive. The Lactoperoxidase System (LPS) recommended by Food and Agriculture Organization for preservation of raw milk is a safe method that can be used in situations where no cooling facility is available or affordable. However, policies to support use of LPS as a method of milk preservation have not been made.

The formal sector which is involved in milk processing, value-addition, increasing shelf-life, and packaging to ensure safety of milk and dairy products is mainly in the hands of public and private milk processors. The sector has the capacity to increase milk intake, processing and packaging to cope with large volumes of milk during the wet season. The high operational costs associated with the formal milk sector are therefore likely to decline as the processors operate at full capacity thus enhancing its efficiency and competitiveness both locally and internationally. However, the strategies for implementation must be participatory involving public and private sectors and relevant stakeholders. In addition, an enabling environment by the government through a legal and regulatory framework and strong institutions will be required to support the development of the sector.

Improved access to support services and technologies

The various farmer co-operatives, self-help groups, private processors, and other partners could be used to provide support services for dairy production in the study area. With improved and expanded road network, there is a great potential to increase access by farmers of essential dairy production services and technologies. Increased use of AI, extension, animal health, training, and credit services will enhance the use of modern farming inputs and appropriate production technologies and hence increase dairy productivity. The improved fodders adapted to the low temperatures and frequent frost bites need to be established in the upper highlands while the suitable fodders for dry areas need to be established in the lower highlands.

The supplementation of dairy stock must be practised judiciously depending on the basal diet offered and the desired level of production. Training of farmers and the other participants who are involved in milk value chain will have a positive impact on adoption of appropriate technologies hence higher dairy productivity. Greater participation of the community and private sector should be encouraged to supplement government efforts to enhance provision of the support services and technologies to farmers. However, better coordination and greater involvement of all key players must be emphasized while the management, accountability, and any investment must be done to the interest of the farmers.

Sustainable use of natural resources

The country’s legislation prohibits grazing, human settlement, deforestation, and crop/ animal production in steep areas, water catchments and wet lands which needs to be enforced for sustainable use of natural resources. Establishment of fodder trees and legumes may improve the fertility status of the soil; hence increased biomass yield in addition to the provision of quality forage for dairy cattle. However, for sustainable protection of the environment and biodiversity conservation, use of natural resources and conservation of fauna and flora in these fragile environments should be done in collaboration and partnership with the relevant stakeholders.

Conclusions

Recommendations

Acknowledgement

The authors appreciate the financial support from the Kenya Government and the International Development Agency (IDA) under the Kenya Agricultural Productivity Project, technical and moral support from the KARI-Naivasha staff, dairy farmers and MoL&FD staff in Nyandarua County.

References

Anon 1992 Assessment of animal agriculture in sub-Sahara Africa. Winrock Institute for Agricultural Development. Morrilton, Arkansas 162.

Baltenweck I, Staal S J, Owango M, Muriuki H, Lukuyu B, Gichugu G, Kenyanjui M, Njubi D, Tanner J and Thorpe W 1998 Intensification of dairying in the greater Nairobi milk-shed: Spatial and household analysis. Ministry of Agriculture (MoA), Kenya Agricultural Research Institute (KARI) and International Livestock Research Institute (ILRI) Collaborative Research Project Report. 26. http://www.smallholderdairy.org/collaborative res reports.htm

Central Bureau of Statistics (CBS) 2001 The 1999 Population and Housing Census. Counting Our People for Development. Volume 1. Ministry of Planning and National Development. Nairobi, Kenya. 14 – 15.

Cherono P K 2005 School milk program for Kenya. Kenya Dairy Board (KDB). Paper presented at the 1^st African School Milk Conference, 28^th September 2005. Kampala, Uganda. 21.

Delgado C, Rosegrant M, Steinfield H, Ehui S and Courbois C 1999 Livestock to 2020: the next food revolution. Food, Agriculture, and the Environment discussion Paper 28. Washington, DC: International Food Policy Research Institute (IFPRI), Food and Agriculture Organization (FAO) and International Livestock Research Institute (ILRI). 83.

Food and Agriculture Organization (FAO) and International Dairy Federation (IDF) 2004 Dairy Development Newsletter. October 2004 – Issue No. 8.

Government of Kenya (GOK) 2007 Kenya Vision 2030. Ministry of Planning and National Development. Government of Kenya. A competitive and Prosperous Kenya. Nairobi, Kenya. 20.

Israel G D 2009 Determining sample size. Program Evaluation and Organizational Development (PEOD), Institute of Food and Agricultural Sciences (IFAS), University of Florida, Gainesville 3261. http://edis.ifas.ufl.edu/pdffiles/PD/PD00600.pdf.

Kenya National Bureau of Statistics (KNBS) 2010 Kenya 2009 Population and housing census results. Ministry of Planning, National Development and Vision 2030. Nairobi, Kenya. United States Agency for International Development (USAID), United Nations Population Fund (UNFPA) and Swedish International Development Agency (SIDA). Nairobi, Kenya. 8.

Kimigo J, Mbuvi J P, Kirochi G and Gicheru P T 2008 Effect of land use change on soil quality in the susumua catchments of Tana River drainage system in Nyandarua District, Kenya. East African Agricultural and Forestry Journal 74 (2): 85-94.

Kiragu J W, Mitaru B N, Siamba D N, Irungu K R G and Ngatia S K 2008 Strategies for early weaning of dairy calves based on forages and solid feeds on growth and rumen function. Paper presented at the 11^th Kenya Agricultural Research Institute (KARI) Biennial Scientific Conference, 10^th – 14^th November 2008, Nairobi, Kenya. 17.

Lanyasunya T P, Rong W H, Mukisira E A and Abdulrazak S A 2006 Performance of dairy cows in different livestock production systems on smallholder farms in Bahati Division, Nakuru District, Kenya. Pakistan Journal of Nutrition 5 (2): 130-134.

Lanyasunya T P, Mukisira E A, Lokwaleput I K and Siamba D N 2001 Factors limiting optimization of smallholder peri-urban dairy herd production in Kenya. Proceeding of the 10^th Conference of the Association of Institutions for Tropical Veterinary Medicine, 20^th-23^rd August 2001, Copenhagen, Denmark. 47-49.

Makenzi P M 2008 Viability of payment for environmental services approach (PESA) as an innovative financing mechanism for conservation of catchments and watersheds: a case study of lake Naivasha and river Malewa watersheds in Kenya. Proceedings of the 3^rd annual Research Week and International Conference, 16^th-18^th September 2008, Agricultural Resource Centre (ARC), Egerton University, Njoro, Kenya. 22.

Mburu L M, Gitu K W and Wakhungu J W 2007 A cost-benefit analysis of smallholder dairy cattle enterprises in different agro-ecological zones in Kenya highlands. Livestock Research for Rural Development. Volume 19, Article # 95. http://www.lrrd.org/lrrd19/7/mbur19095.htm

Ministry of Livestock and Fisheries Development (MoL&FD) 2007 Livestock Production Annual Report 2006. Nairobi, Kenya.

Ministry of Livestock and Fisheries Development (MoL&FD) 2003 Livestock Production Annual Report 2002. Nairobi, Kenya.

Muriuki H, Omore A, Hooton N, Waithaka M M, Ouma M, Staal S J and Odhiambo P 2003 The policy environment in the Kenyan dairy sub-sector: A review. Smallholder Dairy (Research and Development) Project Report No. 2. Nairobi, Kenya. 75. http://www.smallholderdairy.org/collaborative res reports.htm

Mwendia S W, Wanyoike M, Wahome R G and Mwangi D M 2007 Effect of Napier head smut disease on Napier yields and the disease coping strategies in farming systems in central Kenya. Livestock Research for Rural Development. Volume 19, Article #109. http://www.lrrd.org/lrrd19/8/mwen19109.htm

Omore A, Muriuki H, Kinyanjui M, Owango M and Staal S J 1999 The Kenya dairy sub-sector: a rapid appraisal. Small Holder Dairy (Research & Development) Project Report. Ministry of Agriculture (MoA), Kenya Agricultural Research Institute (KARI) and International Livestock Research Institute. Nairobi, Kenya. 51. http://www.smallholderdairy.org/collaborative res reports.htm

Peeler E J and Omore A 1997 Manual of livestock production systems in Kenya. 2^nd Edition. Kenya Agricultural Research Institute (KARI). Nairobi, Kenya. 138.

Romney D L, Njubi D, Njoroge L, Nyangaga J, Waithaka M M, Muriuki H G, Wokabi A, Mwangi D M and Staal S J 2004 Understanding the role of dairy in smallholder systems and the effect of intensification. Proceeding of the 9^th Kenya Agricultural Research Institute (KARI) Biennial Scientific Conference, 8^th-12^th November, 2004, Nairobi, Kenya. 529-535.

SAS Institute Inc. 1999 SAS/STAT Users’ Guide, Version 8 Release 6.03. Statistical Analysis Systems (SAS), Carry, NC. 549-640.

Schreiber C 2000 Sources of innovation in dairy production in Kenya. International Services for National Agricultural Research (ISNAR) briefing paper No. 58. Ministry of Agriculture and Rural Development (MoARD), Kenya Agricultural Research Institute (KARI) and International Livestock Research Institute (ILRI)/ISNAR. The Hague, the Netherlands. 7.

Sombroek W G, Braun H M H and van der Pouw B J A 1982 Exploratory Soil Map and Agro-climatic Zone Map of Kenya. Exploratory Soil Survey Report No. E1, Kenya Soil Survey, Nairobi, Kenya. 56.

Staal S J, Waithaka M, Njoroge L, Mwangi D M, Njubi D and Wokabi A 2003 Costs of milk production in Kenya: estimates from Kiambu, Nakuru and Nyandarua Districts. Smallholder Dairy (Research and Development) Project Report No. 1. Ministry of Livestock and Fisheries Development (MoL&FD), Kenya Agricultural Research Institute (KARI), International Livestock Research Institute (ILRI) and Department for International Development (DFID). Nairobi, Kenya. 30. http://www.smallholderdairy.org/collaborative res reports.htm

Staal S J, Owango M, Muriuki H, Lukuyu B, Njoroge L, Njubi D, Baltenweck I, Musembi F, Bwana O, Muriuki K, Gichugu G, Omore A and Thorpe W 2001 Dairy systems characterization of the greater Nairobi milk shed. Smallholder Dairy (Research & Development) Project. Kenya Agricultural Research Institute (KARI), Ministry of Agriculture (MoA) and International Livestock Research Institute (ILRI) Collaborative Dairy Research Program, Nairobi, Kenya. 73. http://www.smallholderdairy.org/collaborative res reports.htm

Staal S J, Chege L, Kenyanjui M, Kimari A, Lukuyu B, Njubi D, Owango M, Tanner J, Thorpe W and Wambugu M 1998 Characterization of dairy systems supplying the Nairobi milk market. Kenya Agricultural Research Institute (KARI), International Livestock Research Institute (ILRI) and Ministry of Agriculture, Livestock Development and Marketing (MALDM) Collaborative Dairy Research Program. Nairobi, Kenya. 41. http://www.smallholderdairy.org/collaborative res reports.htm

Thorpe W, Muriuki H, Omore A, Owango M and Staal S J 2000 Development of smallholder dairying in Eastern Africa with particular reference to Kenya. Paper prepared for the University of Zimbabwe (UZ), Royal Veterinary and Agricultural University of Denmark (RVAU), Dublin Institute for Advanced Studies (DIAS) and the Danish International Development Agency-The Danish Bilateral Programme for Enhancement of Research Capacity in Developing Countries (DANIDA-ENRECA) Project Review Workshop 10-13 January 2000. Bronte Hotel Harare, Zimbabwe. 10. http://www.smallholderdairy.org/publications/Conference/Thorpe%20et%20al-2000-Dev%20of%20smallholder%20dairying%20in%20EA-Zimbabwe.pdf