Livestock Research for Rural Development 26 (8) 2014 Guide for preparation of papers LRRD Newsletter

Citation of this paper

Water sources and quality for dairy cattle in smallholder farms in semi-arid Kenya

D M G Njarui, J K Itabari, J M Kabirizi1 and A J Mwilawa2

Kenya Agricultural Research Institute - Katumani Research Centre, PO Box 340 – 90100, Machakos, Kenya,
1National Livestock Resources Research Institute, PO Box 96, Tororo, Uganda
2Tanzania Livestock Research INstitute, Mabuki, PO Box 352 Mwanza, Tanzania


The quality and provision of water is not optimal to maximize dairy cattle performance and health in the semi-arid ecosystems of Kenya. A survey targeting 56 smallholder dairy farmers was conducted around 15-km radius of peri-urban areas of Wote town in the eastern region of Kenya. The farmers were interviewed one-on-one using a simple structured questionnaire in May and June 2013. The quality of water from 12 sources was analysed. The objective of the study was to assess the sources and quality of water for dairy cattle and coping strategies among smallholder farmers in peri-urban areas of eastern Kenya.

The study showed that to mitigate water scarcity, the farmers had diversified options of water sources with approximately 57% obtaining water from more than one source. The major sources of water for dairy cattle were shallow wells and valley bottoms. However, the water from shallow wells and valley bottoms had high levels of chlorides (mean 67.77 and 70.37 me/litre, respectively) and sulphates (99.33 and 25.63 me/litre, respectively) and was saline thus not suitable for dairy cattle. The pH was alkaline (pH >8.0) and above the preferred level for livestock. Good quality water was limited mainly to rainwater harvested from houses rooftop and in water pans. Nevertheless, the quantity stored from rooftops was normally low (<5000 litres per household) and was occasionally used for livestock in most farms. It was regarded as premium water due to its high quality and was mainly preserved for household use. Ox-carts were the most widely used (41%) means of transporting water while bicycles were the least used (5.4%).

To supply adequate good quality water for dairy cattle, there is need to harvest more rainwater from rooftops by installing gutter systems and increasing storage capacity by acquiring larger tanks.

Key words: dairy farmers, milk production, peri-urban, water quality, Wote town


Water is an essential element for all life and is important for domestic use, industrial purposes (Lade et al 2012), crop production and livestock farming. Although water constitutes between 60 and 70% of the body of animals, it is often overlooked as an important nutrient for livestock. It is required for digestion and metabolism of energy and nutrients; circulation of nutrients and metabolites to and from tissues; excretion of waste products (via urine, faeces, and respiration); and maintenance of proper ion, fluid and heat balances (Houpt 1984; Murphy 1992). Water intake problems limit milk production of dairy cattle and growth and adversely affect health and productivity.

Water requirement in animals is influenced by productivity, size of animal and environmental conditions. The water requirement per unit of body mass of a high-producing dairy cow is greater than that of any other land-based mammal (Woodford et al 1985). Thus, water requirements of a lactating cow are closely related to milk production, moisture content in the feed and environmental factors such as air temperature and humidity. Total water content of the bodies of adult dairy cattle ranges between 56 and 81% of body weight depending on stage in the lactation cycle (Murphy 1992). Milk is composed of nearly 87% water (Ward 2014) and thus dairy cattle need plentiful water to achieve high milk production. Dry cows need 30 - 50 litres of water daily while lactating cows require between 50-100 litres daily (Lardy et al 2008) depending on amount of milk production.

The semi-arid ecosystems of Kenya are characterised by water scarcity, both in quantity and quality. The region experiences low (annual 500-800 mm) and erratic rainfall. There has been increased drought frequency due to effects of climate change and variability thus exacerbating the water scarcity. In the semi-arid eastern region of the country, about 90% of total annual rainfall falls in two rainy seasons: March to May and mid-October to December. Inter-seasonal rainfall variation is large with coefficient of variation between 45 and 58% (Keating et al 1992). There is a prolonged dry season from June to mid-October. Permanent streams that were major sources of water dried up over 20 years ago due to effect of climate change and what remains are river lines or valley bottoms which act as water ways during the rains but cease to flow within 2 - 3 weeks after the end of rainy season. Consequently, for most of the year, the quality and provision of water is not optimal to maximize dairy cattle performance and health. Water harvesting from surface run-off and storage has been recommended as one option to resolve water scarcity in parts of the world (Frot et al 2008) and is applicable in semi-arid environments.

As the growth of dairy cattle farming in the peri-urban region of eastern Kenya expands due to demand for dairy products from rising population and improved incomes, inadequate water is a key challenge to the growth of the industry, particularly during the dry season. Maintaining plentiful supply of good quality water is essential for growth and development of the sub-sector. There is a need to understand the sources of water, bottlenecks and coping strategies which farmers adopt in order to develop alternative options of water sources and enhance their resilience against climate variability. The objective of the study was to assess the sources and quality of water for dairy cattle and coping strategies among smallholder farmers in peri-urban areas of Eastern Kenya.

Materials and methods

Description of site

The study was conducted in the peri-urban areas of Wote Town, Makueni County in the eastern region of Kenya. Wote Town (37°37’E; 1 °47’S) is situated 135 km south-east of Nairobi City, the capital of Kenya, at 1100 m above sea level. It is a major commercial centre and the administrative headquarters of Makueni County. The peri-urban areas lie in Lower Midland (LM) 4 agro-ecological zone according to Jaetzold et al (2006) and experience semi-arid climate with low (average annual rainfall 500 mm) and unreliable rainfall. The rainfall is bimodal, with the long rains occurring from March to May and the short rains from mid-October to December. There is a long dry season from June to mid-October. Evaporation is high (annua1 evaporation: 1600 - 2300 mm) and exceeds the amount of rainfall (KARI 2001) in all months except November when rainfall is greater than evaporation. The average temperature ranges from 20 to 30°C.

Wote has a population of approximately 56,000 people of whom 5,500 are classified as urban. The population of Makueni County is 885,000 according to a 1990 census (Wanjara 2011), with an area of 8,008 km2. About 80% of farmers are smallholder and practice predominantly mixed crop-livestock farming. On average they keep 5.84.0 dairy cattle per household mainly under semi-intensive production systems (Njarui et al 2012). The dairy cattle kept are the European breeds (Bostaurus): Holstein-Friesian, Aryshires, Guernsey and Jersey and their crosses with local zebu (Bosindicus). The site was selected for the study since there are established smallholder dairy cattle farms.

Sampling and data collection

The sampling procedure has been described by Njarui et al (2012). The survey area covered a 15 km radius around Wote Town. The target population for the study was defined as smallholder farmers having at least one grade (cross between local zebu and exotic European dairy cattle breeds) dairy cow. A proportional stratified sampling method was employed and was based on geographical location of households within the urban centre. The target area was divided into four clusters relative to the town: north, south, east and west. From each direction, a list of farmers with at least a grade dairy cow was compiled with the help of extension officers working in the county. A total of 56 households were selected using simple random sampling and interviewed. Information solicited included sources of water for livestock and means of transport, using a simple questionnaire for in-person interviews and discussions. The distance to water sources was measured and 12 samples of water were taken from roof catchment, shallow well, water pan and valley bottom for quality analysis. Three samples were taken from each source and were strategically collected in different locations so that they were representative of the study area. The samples from valley bottom were taken at different points along Kaiti riverline which is a major source of water in the region. The pH was measured using a procedure based on 1:2.5 soil:water while sodium, chlorides and sulphates contents were determined using the methods of Buurman et al (1996). The survey was conducted in May and June 2013.

Data analysis and presentation

Data was coded and entered in an excel spreadsheet, checked for errors and information on water sources and mean of transport derived using the Statistical Package for Social Sciences (SPSS) version 12 for Windows (SPSS 2002). The pH and salt values were statistically evaluated by analysis of variance using Statistical Analysis Software (SAS) (SAS 1991) and means separated by Least Significant Difference (LSD) (Steel and Torrie 1981). The results are presented using descriptive statistics, tables, graphical illustrations and photographs.

Results and Discussion

Water for dairy cattle

There were differences in the amount of water made available to the dairy cattle among production systems. Overall, 75% of farmers who kept dairy cattle under stall-feeding, provided water ad libitum while 25% did not have water available to the animals throughout the day. For animals under a combination of stall feeding and grazing, only in 66.7% of farms where water was provided ad libitum, while under free grazing only in 40% of farms, animal had access of water throughout the day. The proportion was lower under free grazing because in this system, animals were taken to watering points in the afternoon after grazing while in the other systems water was provided in troughs. Farmers were generally not aware of the quantity of water their animals need to maintain high milk production, hence the reason why in some instances they did not provide water throughout the day. Further, water availability is limited particularly during the dry season.

Water intake depends on milk production of animal among other factors with high producers requiring more water (Woodford et al 1985). Other determinants include dry matter intake, relative humidity and moisture content of feeds. This implies that lactating cows without adequate water will produce less milk than cows having free access to water because the first group does not obtain sufficient water required for milk production. Temperatures are generally high in semi-arid climates and consequently animals would require more water to achieve high milk production. In view of this, there is need to determine the lower and upper limit of water requirement for dairy cattle in these regions.

Sources of water

The major sources of water for dairy cattle were shallow wells and valley bottoms. Other sources included rooftops, boreholes, water pans and piped water (Table 1). Overall, 43.0% of farmers obtained water from two sources while 13.7% had three sources. Shallow wells were usually about 20 - 40 m deep and were lined with concrete to hold the soil (Figure 1). Additionally, in instances where a hand pump was installed, a concrete lid was put on top (Figure 2). Where water was obtained from house rooftops, this was accomplished via gutter systems and water was stored in plastic and or concrete tanks (Figure 3). Although over 95% of dairy farmers in the peri-urban areas of Wote Town harvested water from rooftops, a much smaller proportion used it for livestock albeit not on a daily basis. The water from rooftops was highly valued because it was of high quality and reserved for household drinking and cooking. Further, although most households had potential of harvesting more water from iron sheet rooves, the quantity stored was normally low (<5000 l). This amount was not sufficient to supply all household consumption and livestock requirements. There is need to harness more water from this source in order to improve supply for livestock since it is of good quality.

Where water was obtained from river-lines or valley bottoms, this was by excavating the sand to create a shallow depression for water to percolate from underground and then scooping into jerry cans (Figures 4 and 5). These sources were usually temporary and normally lasted for a few weeks after which a new point was excavated. Piped water was only available in a few households near Wote Town. Water pans were not popular due to their high loss of water through evaporation as a result of high temperatures experienced in semi-arid. The results clearly indicate that to mitigate water scarcity in the region, farmers had diversified their options of water sources with high proportion having more than one source.

Table 1. Sources of water for dairy cattle under different production systems in peri-urban area of Wote Town, eastern Kenya
Sources Production systems Total
Stall- feeding
Stall-feeding and grazing
% of water source
Shallow wells 12.5 16.7 0 13.4
Valley bottoms 12.5 8.3 20 10.9
Boreholes 0 0 40 5.7
Roof tops 25 4.2 0 8.1
Water pans 25 0 0 5.4
Earth dam and valley bottom 0 12.4 20 10.8
Valley bottoms and rooftops 0 25 0 16.1
Shallow wells and rooftops 12.5 20.8 0 16.1
Shallow wells, valley bottoms and rooftops 0 4.2 20 5.6
Valley bottoms, rooftops and piped water 12.5 8.4 0 8.1

Figure 1. A woman draws water from an open,
shallow well lined with concrete
Figure 2. Children draw water from a shallow
well equipped with a hand pump

Figure 3. Rainwater harvested from house rooftop using gutter systems and stored in concrete tanks

Figure 4. Typical valley bottom with sands after the rainy seasons Figure 5. A young man scoops water from a valley
bottom after excavating the sand
Transporting water

The distance to the source of water ranged from 0.2 to 5 km excluding water from rooftop catchment. Most of the shallow wells were individually owned and were normally within the farm, thus distance was short. However, there were a few shallow wells that were dug by non-governmental organisations for public use. The distance of 5 km is considered relatively short, compared with much drier regions of LM 6 agro-ecological zone where water is sometimes sourced from over 20 km. As the distance was variable, farmers used different means to transport water for their dairy cattle. Ox-carts were the most popular means of transport with approximately 41% of households using them, followed by people and donkeys (21.6% each) while bicycles were least used (5.4%) (Figure 6). Bicycles, although important means of transport in semi-arid regions, were not widely used because of the hilly terrains which demand more human power to peddle. Further, they can only transport limited quantity of water (three jerry cans of 20 l each) at a time (Figure 7).

On the other hand, ox-carts were popular because they carry more water and are cheaper than tractors. The carts were either pulled by oxen (Figure 8) or donkeys (Figure 9); households which did not own an ox-cart used donkeys to carry the water . Oxen were kept for multiple purposes: in ploughing land for crop production or sold for cash during crop failure. Donkeys, by contrast, were kept for exclusively for transport because they are resilient to dry conditions and can travel long distances. Wheel barrows and people were mainly used to transport water within a short distance particularly where the source was located within the farm. Males were generally the one who transported water using wheel barrows particularly the youths because are masculine and energetic (Figure 10). On the other hand, where people transported water, it was entirely females using their back to carry water in jerry cans (Figure 11). Where tractors were used, they were normally hired to transport large quantities of water using water bowsers (Figure 12).

Figure 6. Means of transporting water for dairy cattle in peri-urban areas of Wote Town, eastern Kenya

Figure 7. Bicycles were used for transporting water
in peri-urban areas of Wote Town
Figure 8. Oxen transporting water in peri-urban areas of Wote TownOxen
transporting water in peri-urban areas of Wote Town

Figure 9. Donkeys carry water in peri-urban
areas of Wote Town
Figure 10. Young man push wheel barrow loaded with jerry can of water

Figure 11. Woman carry water in 20 l jerry can Figure 12. Tractor pulling a water bowser
Water quality

The pH and salt levels of water from difference sources is given in table 1, the mean values in table 2 and are graphically shown in figure 2 for pH and figure 3 for salts levels.

There were variations in the quality of water from different sources and also from similar sources with large variations in chlorides contents (Table 2). The pH ranged from around neutral (pH 6.58) in water from rooftops to alkaline (pH >8.0) in water from shallow wells and valley bottoms (Figure 13). The salt levels were highest in water from shallow wells followed by water from valley bottoms and lowest in water from rooftops. The mean level of sodium was 24.6 me/l in water from shallow wells while from rooftops it was 0.74 me/l. The mean chloride and sulphate levels were 67.8 and 99.3 me/l, respectively, in water from shallow wells (Table 3; Figure 13). In water from rooftops, the levels were 13.0 and 6.06 me/l for chlorides and sulphates, respectively. The high level of salts in water from shallow wells is attributed to inherent salty nature of basement rocks while in the valley bottoms this is due to high salt levels along the valleys. Due to the high chlorides and sulphates levels, all the water sampled from shallow wells and valley bottoms was saline and not suitable for livestock. This information was corroborated by farmers who observed that livestock reduced milk production when given water from shallow wells particularly during the long dry season. The water was also alkaline with a pH of above 8.0. The preferred pH of drinking water for dairy animals is 6.0 to 8.0 (Olkowski 2009). Low quality water normally reduces water intake and feed consumption in animals (Lardy et al 2008). Specifically, in dairy cattle, waters with a pH outside the preferred range results in reduced milk yield and milk fat, low daily gains, metabolic disorders, and reduced fertility (Olkowski 2009).

Good quality water was limited mainly to rainwater harvested from the house rooftops and to some extent water in natural pans. Since the pH was around neutral and it contained low salt levels, the water can be regarded as most suitable for dairy cattle. Nevertheless, in one of the farms, the water pan contained high level of sulphates and this was attributed to the high salt level along the water path before it emptied into the water pan.

Table 2. Chemical content of water from different sources in peri-urban areas of Wote Town, eastern Kenya
Farmers Village Salt levels Sources of water
(1:2.5 soil:H20)
Philip Masika Kaumoni 6.37 0.09 15.0 0.57
Thomas Muoka Malivani 7.18 2.10 14.5 17.1 Roof water
Vincent Munyoki Mumbuni 6.20 0.02 9.50 0.51
Philip Masika Kaumoni 8.62 31.90 18.8 123
Francisca Ndemange Kitutu 8.46 11.8 14.5 89.7 Shallow well
Alex Ndambuki Malivani 8.15 30.2 170 85.3
Benson Kikolya Unoa 7.86 0.56 11.10 11.2
Philip Masika Kaumoni 6.85 0.21 10.00 1.75 Water pan
Rosina Kyalo Kamunyololo 7.71 0.46 8.90 29.4
Kaiti* Nthangu 8.02 0.97 24.0 28.2
Kaiti* Kasemei 8.35 3.76 36.1 32.3 Valley bottom
Kaiti* Kavati 8.43 1.4 151 16.4
*Water was sampled at different location along the valley bottom

Table 3. Mean pH and salt levels of water from four sources; roof water, shallow well, water pan and valley bottom
Sources of water pH Sodium
Roof water 6.58 0.74 13.0 6.06
Shallow well 8.41 24.6 67.8 99.33
Water pan 7.47 0.41 10.0 14.12
Valley bottom 8.27 2.04 70.4 25.63
SEM 0.24 3.27 32.6 8.07
P 0.002 0.002 0.43 <0.001

Figure 13. Level of pH from four sources of water. Vertical bars represent LSD

Figure 14. Levels of sodium, chlorides and sulphates from four sources of water; roof water,
shallow well, water pan and valley bottom. Vertical bars represent LSD

Conclusions and recommendations

Based on the findings of this study, it was concluded that:


The authors are grateful to smallholder dairy farmers who participated in the study. Special thanks are due to the staff of the Ministry of Agriculture, Livestock and Fisheries Development in Makueni County and the local provincial administrative officers for their support. Our gratitude is extended to the Director, Kenya Agricultural Research Institute (KARI) for supporting this study. This study was funded by Association for Strengthening Agricultural Research in Eastern and Central Africa (ASARECA). The views expressed in this documents are not necessary those of ASARECA.


Buurman P, van Lagen B and Velthorst E J 1996 Manual for Soil and Water Analysis. Backhuys Publishers, Leiden, the Netherlands. pp 324

Frot E, van Wesemael B, Benet A S and House M A 2008 Water harvesting potentials as a function of hillslope characteristics: A case study from the Sierra de gator (Almenria Province, south-east Spain). Journal of Arid Environment 72: 1213-1231

Houpt T R 1984 Water Balance and Excretion. In: M J Swenson (ed) Duke’s Physiology of Domestic Animals. 10th Edition, Comstock Publishing Co., New York.

Jaetzold R, Schmidt H, Hornetz B and Shisanya C 2006 Farm Management Handbook of Kenya. Vol. 2. Natural Conditions and Farm Management Information, 2nd Edition, Part C, Eastern Kenya , Sub-Part C1, Eastern Province.

KARI (Kenya Agricultural Research Institute) 2001 The KARI medium term implementation plan. 1st draft Report. An Agenda of partnership to transform Kenya Agriculture, 2003 -2007. pp 116

Keating B A, Siambi M N and Wafula B M 1992 The impact of climatic variability on cropping research in semi-arid Kenya between 1955 and 1985. In: M E Probert. (ed.)A search for strategies for sustainable dryland cropping in semi-arid Eastern Kenya. ACIAR Proceedings No. 41. Canberra, Australia. pp 16-25

Lade O, Coker A and Sridhar M 2012 Sustainable water supply for domestic use: Application of roof-harvested water for ground water recharge. Journal of Environmental Science and Engineering 1: 581-588

Lardy O, Stoltenow C and Johnson R 2008 Livestock and water. North Dakota State University Extension services. North Dakota, USA. AS-954.

Murphy M R 1992 Water metabolism of dairy cattle. Journal of Dairy Science 75:326-333

Njarui D M G, Kabirizi J M, Itabari J K, Gatheru M, Nakiganda A and Mugerwa S 2012 Production characteristics and gender roles in dairy farming in peri-urban areas of Eastern and Central Africa. Livestock Research for Rural Development. 24.

Olkowski A A 2009 Livestock Water Quality. A Field Guide for Cattle, Horses, Poultry and Swines, University of Saskatchewan. pp 157

SAS 1991 SAS/STAT User’s Guide, Version 6 Edition. SAS Institute Inc., Cary, North Carolina, USA. pp 1028

SPSS 2002 Statistical Package for Social Sciences. SPSS B1 survey tips. SPSS Inc. Chicago, USA.

Steel R G D and Torrie J H 1981 Principles and Procedures of Statistics, Second Edition. McGraw–hill book Company, Auckland, New Zealand. pp 236

Wanjara J 2011 Regional demographic and social economic profile, Eastern South region. National Council for Population and Development. Nairobi, Kenya. pp 44

Ward D 2014 Water requirements of livestock. OMAFRA Factsheet Order No 07-023 Ministry of Agriculture and Food, Ontario, Canada.

Woodford S, Murphy M and Davis C 1985 Why cows need water. Dairy Herd Management 2: 36-40

Received 17 April 2014; Accepted 24 June 2014; Published 1 August 2014

Go to top