Livestock Research for Rural Development 25 (9) 2013 Guide for preparation of papers LRRD Newsletter

Citation of this paper

A comparison of animal-crop mixed farming systems in dry lowland Sri Lanka

U Y N Vithanage, M B P Mahipala*, L H P Gunaratne** and H W Cyril*

Postgraduate Institute of Agriculture, University of Peradeniya, Peradeniya, 20400, Sri Lanka
* Department of Animal Science, Faculty of Agriculture, University of Peradeniya, 20400, Sri Lanka
** Department of Economics, Faculty of Agriculture, University of Peradeniya, 20400, Sri Lanka


This study characterizes existing animal-crop mixed farming systems in dry-lowland Sri Lanka. A rapid rural appraisal was conducted via informal discussions with stakeholders of cattle/buffalo farming systems followed by a structured survey with 96 randomly selected cattle/buffalo farmers using a pre-tested, structured questionnaire. This study distinguishes two cattle/buffalo farming systems, namely, semi-intensive farming system (SIFS) and extensive farming system (EFS).

Farming was the main income-generating source (91.1% of farm income) for the majority of households in the area. Cattle/buffalo, upland crop, paddy and poultry farming were the major farming activities that were undertaken by the farmers in both systems. Socio-economic status in SIFS was mainly determined by paddy farming which was the major enterprise (83.2% of total farm output), whereas in EFS it was contingent on cattle/buffalo farming (63.9%). There was some integration among livestock farming, upland crop cultivation and paddy cultivation in SIFS. However, such integration was not obvious in EFS. Although farmers in SIFS used a relatively high level of external inputs for cattle/buffalo farming, their output was significantly lower than EFS, which operated with a large herd (51.3 and 4.64, in EFS and SIFS, respectively) and low external inputs. However, feed, concentrates and mineral intake of the animals was far less than required in both systems. Consequently, low production performances and the frequent incidence of disease were observed. Although it was not obvious in the two farming systems, proper integration among crop-livestock farming activities could be suggested as a better solution to enhance long term sustainability as well as farmers’ living standards by improving farm production. Selling culled animals for meat purposes contributed significantly to the farm output (30.6% of total farm output) in EFS. If there was crop failure, farmers in EFS compensated for it by selling cattle/buffalo for meat purposes. Therefore, cattle/buffalo acted as insurance for the farm family. Conversely, this was not obvious in SIFS, which consisted mainly of Sinhalese (Buddhists) who strongly discouraged meat production because of cultural and religious beliefs. Due to their financial stability, the younger generation was keen on cattle/buffalo farming in EFS. The economic return of cattle/buffalo farming was a major determinant of famers’ involvement in this activity. There was a positive relationship between farmers’ knowledge and involvement in cattle/buffalo farming. Consequently, farmer training programs could be used as an important tool to enhance farmers’ knowledge, which could then improve the productivity of the systems.

Key words: buffalo, cattle, crop, integration


Agriculture and livestock record 11.2% and 0.84% of the total Sri Lankan Gross Domestic Production (GDP), respectively (Central Bank 2011). Total livestock farmer population in the country is about 0.7 million, with livestock activities contributing 30-60% of their gross farm income (Perera and Jayasuriya 2008). Dairy farming is a major component of the livestock industry of the island (Abeyratne 2007).  Currently, the country is about 15-20% self-sufficient in milk, and depends mostly on imported milk powder (Ranaweera 2007).  The government spent 38,200 million Sri Lankan Rupees (LKR) to import milk and milk products in 2011, which was an increase of 46.3% for the same compared to 2007 (Central Bank 2011). It appears that the dependency on imported milk and milk products will continue to increase in future.

There are four main cattle rearing farming systems in Sri Lanka; Up- and Mid-Country, Coconut triangle, Wet lowland and Dry lowland. Cattle/buffalo are important capital assets for the peasant farmers in the Dry lowland areas (Ibrahim et al 1999). In 2011, the cattle/buffalo population in Dry lowland areas was 1.01 million. This represented 63.2% of the total stock of the cattle/buffalo population of the island. In the same year, Dry lowland areas recorded 117 million L of milk production (45.6% of the domestic milk production) (Central Bank 2012). These areas played a significant role in national milk production.

Dairy farming in Sri Lanka is predominantly a smallholder, mixed crop-livestock farming operation (Bandara 2000) and is a good income source for crop-livestock farmers in a mixed farm system (Moran 2009). The impact of the national-level dairy development program has been meager because of lack of understanding of the ecological, socio-economic and cultural limitations (Zemmelink 1999). Therefore, identification of the characteristics of cattle/buffalo-crop mixed farming systems in Dry lowland will be helpful to increase national cattle/buffalo production. This will ensure long-term sustainability of livestock-keeping in the island. Moreover, understanding the fundamental characteristics of cattle/buffalo farming systems would be supportive for the development of regional dairy production.

Materials and methods

The study was conducted in animal-crop mixed farming systems in Dry lowland farming systems in Sri Lanka. Prior to the structured survey, a rapid rural appraisal (RRA) was conducted through informal discussions with stakeholders of cattle/buffalo farming systems in the area. Discussions were held with leading small- to medium-scale farmers, officers of major large-scale cattle/buffalo farming companies (i.e., CIC Holdings), officers of formal milk collecting networks (i.e., Milk Industries of Lanka Company Ltd. and Nestle Lanka Ltd.), small-scale milk collectors, veterinary surgeons, livestock development instructors and artificial insemination technicians. The survey was then conducted using a pre-tested structured questionnaire. Ninety-six randomly selected cattle/buffalo farmers were interviewed. The questionnaire was designed to collect information on cattle/buffalo management aspects, inputs, outputs, socio-economic situations and involvement of households in cattle/buffalo farming. The parametric data were analysed by two-tail t test and Z test while non-parametric data were analysed by Chi-square and Fisher’s Extract tests using the SAS statistical package (SAS 1998).

Farmers’ knowledge of cattle/buffalo farming was assessed using 10 targeted questions. Responses were ranked using a five point scale (i.e., very poor, poor, moderate, satisfactory, and highly satisfactory). The level of farmers’ knowledge between identified farming systems was compared using a Chi-square test. Feeding (i.e., forage search, leading animals to graze and offering feed), cleaning (i.e., animals, sheds and milking tools), milking and milk selling were some of the important involvements in cattle keeping (Mahipala and Gunaratne 2003). Involvement of each member of the farm family in cattle/buffalo farming was ranked using a three point scale (i.e., low - no contribution; moderate - engage in one or more but not all activities; high - engage in all activities). Ownership of vehicles and condition of farmers’ dwellings were considered as criteria in assessing the wealth of the farmer. The economic value of the household vehicles (i.e., push bike, motor bike, two-wheeled tractor and four-wheeled tractor) and their houses were ranked using a four point scale. A scatter plot was drawn between the total score of wealth and frequency using Minitab software (MINITAB 2010). Accordingly, three wealth categories were identified as better off, moderate and worst off.


The dynamics of productivity and sustainability of the dairy farming systems are determined by interaction among sub-systems (i.e., cattle/buffalo farming, upland crop farming, paddy farming and poultry farming), and the social structure of the systems. Therefore, efforts were made to identify those dynamics.

Semi-intensive (SIFS) and extensive cattle/buffalo farming systems (EFS)

In addition to dairy farming, cattle/buffalo farmers engaged in paddy (95.8%) and upland crop (56.3%) cultivation.  Further to this, about 33% of cattle/buffalo farmers reared poultry. Grazing and tethering were the principal methods of forage feeding for cattle/buffalo. Tethered animals were fed in stalls with mineral and concentrate supplementation. Consequently, cattle/buffalo farming systems in the area could be reasonably classified into two systems; as semi-intensive farming system (SIFS) and extensive farming system (EFS).

Figure 1 shows the relative importance (i.e., in terms of value of production) and interactions of activities of sub-systems of the SIFS. In the system, socio-economic activities of the farmers were mainly associated with paddy farming. Therefore, cattle/buffalo farming, which contributed only 13.1% of the total output, was the second important enterprise. There was an interaction between cattle/buffalo farming and crop farming. Manure was used to fertilize crops while crop residues were fed to the animals during the harvest season. Feeding cattle/buffalo with straw was common among many farms in the system. During the fallow period, the animals were tethered in paddy fields. The contribution from poultry to total farm output was more than that of upland crop farming.

Figure 1: Schematic representation of the semi-intensive farming system; the relative importance of the sub-activities (indicated by the boxes) is shown in terms of area.
Inputs of the system are indicated by the circles (CC-crop cultivation, F&C-fertilizer and chemicals, HL-hired labour, CF-cut and fed green feed,
TF-tethering green feed, C-concentrate, M- minerals, I-inputs for crop farming like seeds and fertilizer).

Figure 2 schematically shows the relative magnitude (i.e., in terms of value of production) and relationships of sub-systems of the EFS. Inputs used for cattle/buffalo farming were negligible. Therefore, it can be identified as a “low external input agricultural system”. Cattle/buffalo farming was the main constituent while paddy has often been the secondary source of farm income. Consequently, cattle/buffalo farming was more influential on farmers’ socio-economic status. Poultry rearing and upland crop farming were not prominent in the system; both accounted for just 2.43% (1.79% + 0.64%, Figure 2) of the total farm output. Cow dung and urine were not used as a fertilizer for either paddy or upland crop cultivation except when the animals were grazed during the fallow period. Feeding of straw or crop residues to cattle/buffalo was not evident in the system.

Figure 2: Schematic representation of the extensive farming system; the relative importance of the sub-activities (indicated by the boxes) is shown in terms of area.
Inputs of the system are indicated by the circles (CC-crop cultivation, F&C-fertilizer and chemicals, HL-hired labour, GF-grazing feed,
C-concentrate, M- minerals, I-inputs for crop farming like seeds and fertilizer).

Cross-bred and non-descript cattle/buffalo were found in both systems. The percentages of cross-bred cattle, cross-bred buffalo, non-descript cattle and non-descript buffalo were 24.2, 0.88, 54.1 and 20.8, respectively. Herd size of cattle/buffalo farms was significantly (p<0.05) greater in the EFS than the SIFS (Table 1). In terms of cow units (CU; 1CU=300 kg), milking cows, dry cows, heifers, calves and bulls were significantly (p<0.05) greater in the EFS than the SIFS.

Table 1: Herd composition of cattle/buffalo in farming
  Extensive farming systemA Semi-intensive farming systemA
Herd size
Total number of cattle/buffalo 51.3±44.0 4.64±3.57
Total cow unitsa (CU) 44.1±41.4 3.72±2.82
Herd composition in CU
Cows in lactation 15.4±10.6 1.22±1.41
Cows in lactation but not milking 0.668±2.13 0.387±1.08
Milking cowsa 14.7±10.8 0.828±1.08
Dry cowsa 8.96±8.37 0.749±1.00
Heifersa 6.39±8.89 0.607±0.912
Calvesa 7.17±5.04 0.823±0.912
Bullsa 6.22±16.3 0.334±0.753
Lactating/Milking/Dry cow = 1CU; Heifer = 0.75 CU; Calf = 0.50 CU; Bull = 1.50 CU
AMean±Standard deviation
aMeans are different at p<0.05

Some cows in lactation were found to be not milked in either system. There was no difference (p>0.05) between the systems. Monthly herd milk production and monthly meat production were significantly (p<0.05) greater in the EFS compared to the SIFS (Table 2). However, monthly milk production per CU was higher in the SIFS than in the EFS.

Table 2: Production performances of cattle/buffalo farming
  Extensive farming systemA Semi-intensive farming systemA
Milk production
L/herd/montha 782±556 131±138
L/cow unit (CU)/month 13.1±9.66 21.5±23.3
L/cow/day 2.26±2.34 3.42±2.79
Meat production
kg/herd/montha 190±185 8.62±18.9
Possible amount of manure for collection
kg/herd/montha 3620±3600 370±268
Lactating/Milking/Dry cow = 1CU; Heifer = 0.75 CU; Calf = 0.50 CU; Bull = 1.50 CU
AMean±Standard deviation
aMeans are significantly different at p<0.05

Farmers were not eager to collect manure. Therefore, the potential amount of manure collected was predicted based on farmers’ experience, and it was significantly higher (p<0.05) in the EFS than in the SIFS. Significant differences (p<0.05) were observed in feeding practices between the studied systems. The amount of fodder offered and concentrate supplement provided to cattle/buffalo were greater in (p<0.05) the SIFS compared to the EFS (Table 3). Mineral supplements were given to cattle/buffalo only in the SIFS.

Table 3: Feeding cattle/buffalo in farming systems    
  Extensive farming systemA Semi-intensive  farming systemA
Hours allowed for grazing or tethering/daya 13.6±7.64 9.50±2.46
Amount of fodder offered

kg/cow unit (CU)/daya

0.459±1.35 15.4±18.1


8.52±30.7 39.4±56.0
Time allocated for fodder search


0.0298±0.0987 0.582±0.663


0.412±1.25 1.27±1.15
Amount of concentrate feed


0.263±1.35 15.7±29.5


2.07±10.8 72.1±161
Amount of mineral feed


0.00±0.00 0.139±0.251


0.0313±0.161 0.531±1.24
Lactating/Milking/Dry cow = 1CU; Heifer = 0.75 CU; Calf = 0.50 CU; Bull = 1.50 CU
AMean±Standard deviation
aMeans are significantly different at p<0.05
Social structure and level of involvement in cattle/buffalo farming systems

Demographic profile, level of education, occupation, knowledge of cattle/buffalo farming and level of wealth determined the social structure of farming systems. Depending on the time allocated for farming (i.e., paddy, crop, cattle/buffalo and poultry) the farmer (i.e., who devoted more time to farming) and the helper were identified. Associations were evident between farming system and age, education, main occupation, knowledge of cattle/buffalo farming and the level of wealth (p<0.05) (Table 4).

In the EFS, the majority of the farmers belonged to the age category of 15-30 years, whereas most of the farmers in the SIFS were in the age category of 31-55 years. In both systems, the majority of the farmers and helpers were found to be educated up to secondary school level. However, they did not have satisfactory knowledge of cattle/buffalo farming (i.e., very poor or poor). In the EFS, the majority of the farmers did cattle/buffalo farming as their main occupation though they were involved in crop cultivation. None of them were employed by the government. In both systems, the majority of the farmers were found to be moderately wealthy. Furthermore, none of them were ‘better off’ in the SIFS.

Table 4: Comparison of social structure between two farming systems
Variables Frequency (%)
Extensive farming system Semi-intensive farming system
<15 years 0.00 0.00
15-30 years 47.7 15.0
31-55 years 45.5 65.4
> 55 years 6.82 19.7
No formal education 2.27 6.30
Primary education (<5 grade) 15.9 8.66
Secondary education (5-11 grade) 75.0 78.7
Upper secondary education (12-13 grade) 6.82 6.30
Main occupationa
Cattle farming 51.2 1.56
Crop cultivation 37.2 85.2
Farm labour 4.65 1.56
Government employee 0.00 0.78
Security forces 2.33 3.13
Other 4.65 7.03
Knowledge of cattle/buffalo farminga
Very poor 14.8 7.25
Poor 63.0 53.6
Moderate 18.5 27.5
Satisfactory 0.00 11.59
Highly satisfactory 3.70 0.00
Level of wealtha
Worst off 38.5 34.9
Moderate 57.7 65.2
Better off 3.85 0.00
aThere is an association at p<0.05  

There was an association among involvement in cattle/buffalo farming and the variables of age, main occupation, education and wealth in dairy farming in the EFS; and the variables of age, main occupation, education, knowledge and wealth in dairy farming in the SIFS (p<0.05) (Table 5). In the EFS, the younger generation was highly involved in cattle/buffalo farming, whereas older farmers were involved more in the SIFS. It was obvious that farmers who undertook cattle farming as their main occupation were engaged more in cattle/buffalo farming in the EFS. However, it is interesting to note that the SIFS farmers who gave farm labour as their main occupation were involved more in cattle/buffalo farming than the other farmers who cited cattle/buffalo farming as their main occupation. In the SIFS, the involvement in cattle/buffalo farming correlated positively with the farmers’ education. However, such a relationship could not be observed in the EFS. Further to this, in both farming systems, there was an increasing trend of involvement in cattle/buffalo farming in parallel with the farmers’ knowledge.

Table 5: Level of involvement in cattle/buffalo farming
Variable Frecuency (%)
Extensive farming system Semi-intensive farming system
Low Moderate High Low Moderate High
15-30 years 52.4 0.00 47.6 10.5 36.8 52.6
31-55 years 20.0 15.0 65.0 15.7 36.1 48.2
> 55 years 66.7 0.00 33.3 12.0 16.0 72.0
Main occupationa
Cattle farming 18.2 4.55 77.3 0.00 50.0 50.0
Crop cultivation 62.5 6.25 31.3 13.9 32.4 53.7
Farm labour 0.00 50.0 50.0 0.00 0.00 100.0
Government employee 0.00 0.00 0.00 0.00 100.0 0.00
Security forces 100.0 0.00 0.00 0.00 25.0 75.0
Other 50.0 0.00 50.0 30.0 30.0 40.0
No formal education 0.00 0.00 100.0 12.5 62.5 25.0
Basic education (<5 grade) 42.9 14.3 42.9 0.00 72.7 27.3
Primary education (5-9 grade) 36.4 9.09 54.6 18.5 24.6 56.9
Ordinary level (10-11 grade) 54.6 0.00 45.5 8.57 31.4 60.0
Advanced level (12-13 grade) 0.00 0.00 100.0 25.0 12.5 62.5
Knowledge of cattle/buffalo farming
Very poor 25.0 0.00 75.0 20.0 20.0 60.0
Poor 17.7 0.00 82.4 18.9 8.11 73.0
Moderate 20.0 0.00 80.0 0.00 15.8 84.2
Satisfactory 0.00 0.00 100.0 0.00 25.0 75.0
Wealth of the farmera
Worst off 10.0 0.00 90.0 17.4 4.35 78.3
Moderate 26.7 0.00 73.3 6.98 18.6 74.4
Better off 0.00 0.00 100.0 0.00 0.00 0.00
a There is an association at p<0.05


Semi-intensive farming system (SIFS) and extensive farming system (EFS)

Dry lowland played a great role in paddy production in Sri Lanka by producing about 62.6% of the national paddy production in maha seasons in 2010 and 2011 (Department of Census and Statistics 2011). Almost all the farmers (95.8%) were engaged in paddy farming. Additionally, 56.3% and 33.3% of them undertook upland crop cultivation and poultry keeping, respectively. A greater proportion of the household income (91.2%) was generated through farm products. Consequently, it is reasonable to state that households in the study area mainly relied on farming for their income. Ecological and socio-economic factors of the dry zone varied from one place to another. Therefore, differences among farming systems could be observed. Ratnayaka et al (1992) found two cattle farming systems in the dry lowlands; Dry Zone cattle farming System (DZS) where there was no access to water from irrigation schemes and Irrigation Settlements cattle farming Systems (ISS) where there was access to water from irrigation schemes. Farmers in the DZS adapted extensive management of animals for milk and meat production and for draught purposes, whereas farmers in the ISS adapted intensive or tethering management. Non-descript local breeds were reported in both systems while exotic Zebu cross-bred animals were reported only in the ISS.  The reported herd size in the DZS and the ISS was 18.3 and 6.6, respectively. According to the present study, two farming systems, the EFS and the SIFS, were identified in the dry lowlands. These were distinguished by the management system. The specific features of the EFS were more or less similar to those of the DZS, whereas the SIFS was more related to the ISS. The main features of the EFS were that it was managed extensively, mainly for meat production, low milk productivity, a low external agriculture production system and a comparatively higher herd size (51.3±44.0). Therefore, it was mostly analogous to the DZS. The SIFS could be distinguished by tethering management, primarily for milk production, comparatively high milk productivity, high external input usage and a comparatively lower herd size (4.64±3.57). Consequently, it was similar to the ISS.

In the EFS, animals were allowed free grazing in forests, fallow paddy fields and the road/stream/tank sides. Milked animals were fed on grazing lands during day time. Large herds were kept in paddocks during the night. To separate calves from adults, the animals were either segregated in the same paddock or were put in a different paddock. There was no evidence of evening milking, which had the potential to improve milk yield (Hale et al 2003). Animals in the SIFS were tethered in fallow paddy fields and abandoned on uncultivated uplands during the day time. After morning milking, cows were tethered for feeding and brought back late in the evening for evening milking. During the day time, tethering locations were changed one to three times.

In dry lowlands, cattle/buffalo herds were predominantly indigenous Zebu, although Sahiwal cross-bred cattle and improved buffalo were becoming more common in the eastern dry zone, especially in the Mahaweli irrigated areas (Ibrahim et al 1999). The results of the present study are in agreement with the foregoing findings. In the studied area, around three quarters (74.9%) of total cattle/buffalo were non-descript. Inadequately organised breeding programs, poor government services and free grazing, especially in the EFS, were the major constraints to genetic improvement. Cattle/buffalo farmers in the SIFS were more interested in genetic improvement of their animals and, therefore, had been practicing AI over the past 10 years.

According to the current study, the average herd size was 17.8 in the dry lowland. Furthermore, the average herd size in the dry zone varied between 10 and 25 (Ranaweera 2007). Herd size was 11 times higher in the EFS than the SIFS (Table 1). Free grazing in abandoned lands was practiced in the EFS, whereas tethering and night feeding in stalls were practiced in the SIFS as feeding methods. Therefore, the SIFS had a comparatively high labour requirement per CU for tethering and forage searching for stall feeding. Consequently, small herds, which could be handled by family labour, were managed in the SIFS. Due to different reasons like low milk yield, rearing animals for meat, old age, and keeping animals for traditions which were passed down over generations, some cows in lactation were not milked in either system. The percentage of cows, which is the most important category of a dairy cattle herd, should be at least 40% in all zones (Ibrahim et al 1999). The percentage of cows recorded in the present study area was 55.2% and 53% in the EFS and the SIFS, respectively. The results revealed that the farmers managed their herd composition in an appropriate way.

Monthly herd milk production was about six times higher in the EFS compared to that of the SIFS due to larger herd sizes and a higher number of milking cows. However, milk productivity of cows was greater in the SIFS than the EFS as it consisted mainly of cross-breeds (as the genetic potential had been improved by AI in animals which were tied up in the SIFS). Milk productivity of local animals was less than one litre per day but two to five times more in cross-bred animals (Premarathne and Premalal 2005). Moreover, farmers in the SIFS practiced night feeding and concentrate feeding about 33 and 60 times more respectively, compared with the EFS. Management practices of the SIFS allowed genetic improvement of the herds via AI. Greater genetic potential and better nutrition had, therefore, contributed to higher productivity of cattle/buffalo in the system, compared to the EFS.

In the EFS, there was an owner and a herdsman on some farms. At the end of each year, the owner earned a large sum of money by selling animals for slaughter. The herdsman looked after the herd throughout the year and milked in order to fulfill his daily necessities. Therefore, a higher fluctuation in milk production over the year could be observed in the EFS. Devendra and Thomas (2002) showed that livestock management was closely related to the social, cultural and religious lives of the farmers and this could be observed in both farming systems as well. Meat production was more prominent in the EFS than the SIFS. The EFS consisted mainly of Muslims, which had a cultural influence on the farming activities in the production system. In the SIFS, the majority of the inhabitants were Sinhalese (Buddhists) and, consequently, there was a cultural and religious barrier towards meat production. Moreover, in the EFS, income was mainly derived from sales of surplus animals for meat, and milk was a by-product of the production system (Abeyrathne 2007). Therefore, farmers in the EFS earned a considerable profit by selling animals for slaughter. Monthly per household profit from cattle/buffalo farming in the EFS was 6720LKR, whereas the SIFS ran at a loss (-2200LKR). Further, monthly farm savings in the EFS were 23700LKR, whereas in the SIFS they were -2390LKR. Consequently, farmers in the SIFS had no economic investing power and usually borrowed all necessary inputs except seed paddy for paddy farming. At the end of each cultivating season, they were in considerable debt, and were even borrowing money to meet their household requirements. As the harvesting season ended even when the yield was high, they could not receive a reasonable price per unit of paddy due to intervention of the middlemen. This created a debt as expenditure exceeded revenue. They met that economic gap by selling cattle/buffaloes. The cattle/buffaloes, therefore, acted as insurance for the farmers. Moreover, animals acted as a cash buffer and capital reserve in a crop-animal integration system and they were confirmed security for the production system (Devendra and Thomas 2002b).

In both farming systems, it was difficult to collect manure during day time as animals were in the field, but during night time it could be collected. There was a high possibility of using that manure in crop cultivation, but still the farmers in the EFS did not do this. Crop residues were rarely given to the animals. Devendra et al (2000) revealed that significant crop-animal interaction could occur in farming systems. However, such major integration had not happened in the EFS. Nevertheless, animal-crop interaction could be observed in the SIFS to some extent. Proper animal-crop integration will increase productivity and ensure sustainability of the farming systems (De Jong et al., 1994; Devendra and Thomas, 2002; Paris, 2002). In the SIFS, cattle/buffaloes were in the homestead during night time but in the EFS they were in the forest near the homestead. Therefore, the method of animal-crop integration is different in both farming systems (Figures 3 and 4).

Figure 3: Proposed crop-animal integration for the semi-intensive farming system

Figure 4: Proposed crop-animal integration for the extensive farming system

Cattle/buffalo in the EFS mainly depend on available feeding materials in the forest. However, farmers in the SIFS have a potential to provide a considerable amount of fodder. Tree leaves, especially Jack (Artocarpus heterophyllus) leaves, could be used for feeding goats in both systems, and legume fodder like Gliricidia, coming from living fences, is a good nitrogen source and improves the quality of feed as well. Crop residues are available mainly in two seasons (the harvesting period of the yala and maha) a year. Paddy straw should be stored in a dry place to use in off-seasons, especially during drought periods when natural forage is limited. Animal manure could be used as a fertilizer for paddy and upland and lowland crops in both farming systems. Higher production of animal manure in the EFS ensured lower fertilizer costs by reducing synthetic fertilizer application compared to the SIFS. The introduction of home compost preparation units would enhance the nutrient recycling process in both systems. Moreover, goats should be introduced to both farming systems to increase farm productivity by maximizing the use of farm resources. An awareness program on crop-animal integration should be conducted particularly for the EFS. Similar recommendations were suggested for North Western Province of Sri Lanka by Ibrahim and Jayatileka (2000).

In the dry zone, there are 400,000 hectares of natural grassland (Ibrahim et al 1999). Therefore, almost all the farmers rely mainly on outside own-farm natural grassland feed resources. Ibrahim and Zemmelink (2000) revealed that access to outside own-farm fodder resources is very important for economic dairy production. However, the production performance indicated that the feed intake was insufficient to fulfill their nutrient requirement, especially during the dry period, in both systems. Also, lack of good quality feed production throughout the year was a major constraint for profitable dairy production faced by the small-hold dairy farmers (Chandrasiri 2002). In both farming systems, some farmers practiced stall feeding during night time. They collected grass and fodder from road/stream/tank sides, fallow paddy fields, bunds of paddy fields, live fences and so on.

However, the amount of daily fodder offered per CU was negligible when considering the daily fodder requirement of the animals, resulting in low performances. This is consistent with the finding of Hitihamu et al (2007) that inefficient feeding management contributed to a drop in the supply of the necessary nutrition for high-level production. Most of the naturally available grasses were low in digestibility, less in crude protein content and lower in dry matter. Most of the farmers in the EFS practiced a low external input agricultural system. Farmers in the EFS did not practice mineral feeding, whereas the farmers in the SIFS practiced it to a small extent. Poonac, rice bran and formulated feed were used as concentrates. However, concentrate feeding was far below their requirement, and this also led to low production performance of the animals. As well, Hitihamu et al (2007) stated that farmers in the dry zone mainly practiced an extensive management system and cattle were not fed with concentrate feeds. According to Dillon et al (1997), there was a significant (P < 0.01) linear increase in milk yield as the concentrate feeding level increased, while milk protein yield increased with the level of concentrate. Supplementary feeding with concentrates significantly (P < 0.05) reduced body-weight loss and increased body-weight gain. Concentrate supplementation significantly (P < 0.05) reduced the number of services per conception.

Social structure and level of involvement in cattle/buffalo farming systems

The younger generation of the population was highly engaged in cattle/buffalo farming in the EFS compared to the SIFS. If there was a paddy crop failure in the SIFS, it affected the household economy severely because paddy was the main source of income, while altogether other farming enterprises contributed only about 16.8%. However, in the EFS cattle/buffalo farm production contributed more than one and half times more. Consequently, even though there was a crop failure it did not affect the household economy drastically. Because of that, the EFS had security in economic terms. Therefore, the younger generation in the EFS was engaged more in cattle/buffalo farming, and this is a sign of future sustainability. Because of lower returns generated by cattle/buffalo farming in the SIFS, farmers who were doing cattle/buffalo farming as their main occupation were involved to a lesser extent in cattle/buffalo farming. However, cattle/buffalo farmers in the EFS were involved more in cattle/buffalo farming, and their actions were always reinforced by the relatively high income generated by cattle/buffalo farming.

In both farming systems, the majority of farmers and helpers had secondary education and their knowledge of cattle/buffalo farming was unsatisfactory. Moreover, involvement in cattle/buffalo farming increased with farmers’ knowledge of cattle/buffalo farming. Therefore, to enhance the productivity of the system, training programs on cattle/buffalo farming would be beneficial. In both farming systems, the majority of the farmers were moderately wealthy, and most of them had the expectation of a better life. Furthermore, the majority of Sri Lankan farmers were mainly dependent on subsistence agriculture and were normally poor (De Silva and Ratnediwakara 2010). Improved smallholder livestock practices (Ahmed 2000; Rao and Birthal 2008), especially crop-livestock integration (Moran 2006), are important for alleviating poverty by improving farm production. In the EFS, farmers were mainly engaged in cattle/buffalo farming but in the SIFS, they mainly cultivated crops. Therefore, crop-livestock integration had the potential to improve farm production by using existing farm resources. Moreover, mixed farming (including crop and livestock) systems could be identified as an efficient approach to improve farm production (Ibrahim and Schiere 2002).



The authors gratefully acknowledge the invaluable support given by Dr. (Ms.) H.L.J. Weerahewa, dairy farmers, Officers from Milk Industries of Lanka Company Ltd., Nestle Lanka Ltd., CIC holdings, veterinary surgeons, livestock development instructors and artificial insemination technicians who have given their invaluable support to make this study a success.


Aberathne A S 2007 A review of livestock industry of Sri Lanka, past performance and future trends. Kandy Printers (Pvt) Ltd., Sri Lanka.

Ahmed M M, Jabbar M and Ehui S 2000 Household-level economic and nutritional impacts of market-oriented dairy production in the Ethiopian highlands. Food & Nutrition Bulletin Food and Nutrition Bulletin 21(4): 460-465: Retrieved December 11, 2012, from

Bandara B H W M U S 2000 The current status of smallholder dairy systems in Sri Lanka. Department of Animal Production and Health, Gatambe, Peradeniya, Sri Lanka: Retrieved December 01, 2012, from

Central Bank of Sri Lanka 2011 Annual report. Central Bank of Sri Lanka, Colombo, Sri Lanka: Retrieved September 17, 2012, from

Central Bank of Sri Lanka 2012 Economic and Social Statistics of Sri Lanka 2012. Central Bank of Sri Lanka, Colombo, Sri Lanka: Retrieved January 05, 2013, from

Chandrasiri A D N 2002 Country report on the state of animal genetic resources (Sri Lanka). Food and Agriculture Organization, Rome, Italy: Retrieved November 18, 2012, from

De Jong R, Kuruppu L G, Jayawardena Q W and Ibrahim M N M 1994 Performance of small scale livestock/crop demonstration-cum-training farms in Sri Lanka. Asian-Australasian Journal of Animal Sciences 7(4): 571-582.

De Silva H and Ratnediwakara D 2010 ICT policy for Agriculture based on a transaction cost approach: some lessons from Sri Lanka. International Journal of ICT Research and Development in Africa 1(1): 51-64.

Department of Census and Statistics 2011 Paddy statistics. Department of Census and Statistics, Colombo, Sri Lanka: Retrieved October 09, 2012, from

Devendra C and Thomas D 2002 Crop–animal interactions in mixed farming systems in Asia. Agricultural Systems 71: 27-40.

Devendra C and Thomas D 2002b Crop–animal systems in Asia: importance of livestock and characterisation of agro-ecological zones. Agricultural Systems 71: 5–15.

Devendra C, Thomas D Jabbar M A and Zerbini E 2000 Improvement of Livestock Production in Crop–Animal Systems in Agro-ecological Zones of South Asia. International Livestock Research Institute, Nairobi, Kenya.

Dillon P S, Crosse and Brien B O 1997 Effect of concentrate supplementation of grazing dairy cows in early lactation on milk production and milk processing quality. Irish Journal of Agricultural and Food Research 36: 145-159.

Hale S A, Capuco A V and Erdman R A 2003 Milk yield and mammary growth effects due to increased milking frequency during early lactation. Journal of Dairy Science 86(6): 2061-2071.

Hitihamu H M S J M, Samantha N P G and Rathnayaka R M G K B 2007 Factors Affecting the Cost of Production of Milk in the Kurunegala District. Hector Kobbekaduwa Agrarian Research and Training Institute, Colombo 07, Sri Lanka.

Ibrahim M N M and Zemmelink G 2000 A comparative evaluation of integrated farm models with the village situation in the forest-garden area of Kandy, Sri Lanka. Asian-Australasian Journal of Animal Sciences 13(1): 53-59: Retrieved December 14, 2012, from

Ibrahim M N M and Schiere J B 2002 Introduction to Agricultural System Studies. Faculty of Agriculture, University of Peradeniya, Sri Lanka.

Ibrahim M N M and Jayatileka T N 2000 Livestock production under coconut plantations in Sri Lanka: Cattle and buffalo production systems. Asian-Australasian Journal of Animal Sciences 13(1): 60-67.

Ibrahim M N M, Stael S J, Danniel S L A and Thorpe W 1999 Appraisal of the Sri Lanka Dairy Sector. Faculty of Agriculture, University of Peradeniya, Sri Lanka: Retrieved November 23, 2012, from

Mahipala M B P and Gunaratne L H P 2003 Comparison of cattle farming systems in up country wet zone: cattle management, social structure and involvement in cattle farming. Sri Lankan Journal of Animal Production 2: 41-53.

MINITAB 2010 MINITAB Inst., Inc., USA.

Moran J 2006 Business management for tropical dairy farmers. Landlinks Press, CSIRO, 150 Oxford Street, Collingwood VIC 3066, Australia: Retrieved October 17, 2012, from

Moran J B 2009 Key performance indicators to diagnose poor farm performance and profitability of smallholder dairy farmers in Asia. Asian-Australasian Journal of Animal Sciences 22(12): 1709-1717.

Paris T R 2002 Crop–animal systems in Asia: socio-economic benefits and impacts on rural livelihoods. Agricultural Systems 71: 147-168.

Perera B M A O and Jayasuriya M C N 2008 The dairy industry in Sri Lanka: current status and future directions for a greater role in national development. Journal of National Science Foundation Sri Lanka 36: 115- 126: Retrieved January 07, 2013, from

Premarathna S and Premalal G G C 2005 Country Pasture/Forage Resource Profile. Food and Agriculture Organization, Rome, Italy: Retrieved December 18, 2012, from

Ranaweera N F C 2007 Improved Market Access and Smallholder Dairy Farmer Participation for Sustainable Dairy Development. Central Bank of Sri Lanka, Colombo, Sri Lanka: Retrieved July 23, 2012, from

Rao P P and Birthal P S 2008 Livestock in Mixed Farming Systems in South Asia. International Crops Research Institute for the Semi-Arid Tropics, Patancheru, Andhra Pradesh, India: Retrieved September 24, 2012, from

Ratnayaka D, Abegunawardena H, Jayatilaka M W A P and Abewansa W D 1992 Identification of distinctive cattle farming systems in Sri Lanka and estimation of their production costs. Tropical Agricultural Research 4: 225-245.

SAS 1998 SAS/STAT user’s guide. Version 9.1. SAS Inst., Inc., Cary, N.C.USA.

Zemmelink G, Premaratne S, Ibrahim M N M and Leegwater P H 1999 Feeding of dairy cattle in the forest-garden farms of Kandy, Sri Lanka. Tropical Animal Health and Production 31(5): 307-319.

Received 2 May 2013; Accepted 28 August 2013; Published 4 September 2013

Go to top