The significance of predation as a constraint in scavenging poultry systems: some findings from India

C Conroy, N Sparks^*, D Chandrasekaran^**, A Sharma^***, D Shindey^***, L R Singh^***, A Natarajan^** and K Anitha^**

Natural Resources Institute, University of Greenwich,
Central Avenue, Chatham Maritime, Kent ME4 4TB
m.a.conroy@gre.ac.uk
^*Avian Science Research Centre, Animal Health Group, Scottish Agricultural College, West Mains Road, Edinburgh, EH9 3JG
^**Veterinary College and Research Institute, Namakkal 637 001, Tamil Nadu, India
^***c/o BAIF Development Research Foundation, Dr. Manibhai Desai Nagar, N.H. 4, Warje, Pune 411 029, India

Abstract

 

A research project has been investigating the production problems facing backyard poultry-keepers in two locations in rural India, Udaipur district in Rajasthan and Trichy District in Tamil Nadu. Backyard poultry-keeping is a significant livelihood activity for many poor rural families in India, and for women in particular. A baseline survey of 90 backyard poultry-keepers provided a general overview of socio-economic factors, practices and constraints.

 

When asked about their main constraints, poultry-keepers ranked predation as being as important as, or more important than, disease. The baseline survey collected some quantitative data that showed high mortality rates in chickens in both locations, primarily due to predation. The findings of the baseline survey regarding the relative importance of predation were subsequently confirmed by a one-year monitoring programme. Other papers on constraints to scavenging poultry in India mention predation, but do not identify it as the principal constraint: possible reasons for this are discussed.

Keywords: Chickens, constraints, India, predation, scavenging poultry

 

Introduction

 

There has been relatively little research in India on village chickens, regarding both constraints and technological improvements that could be affordable to the resource-poor. Instead, research (much funded by commercial producers) has focused on intensive production systems. What limited research there has been on scavenging poultry has focused primarily on 'improved' breeds, as was reflected in several papers presented at a national seminar in December 2002 (Devegowda et al 2002).

 

A research project managed by the Scottish Agricultural College, and with socio-economic inputs from the Natural Resources Institute, has been making a modest contribution to filling the research gap by looking at other aspects of improving scavenging systems. The project, which began in late 2000, is funded by DFID's Livestock Production Programme. The locations, both semi-arid, are Udaipur district in Rajasthan and Trichy District in Tamil Nadu; and BAIF Development Research Foundation and Tamil Nadu University of Veterinary and Animal Sciences (TANUVAS) are the collaborators in the respective districts.

 

The two project locations are quite different as far as poultry-keeping is concerned. In the Udaipur project villages the local people are primarily poor tribals, and there is no organised market for chickens. By contrast, in Trichy the poultry-keepers belong to a range of castes and wealth categories. Chickens from this area are highly prized for their superior taste. There is a well-developed commercial market, with traders visiting some of the villages and local markets to purchase birds for sale in urban centres 30-150 kms away.

 

 

Project Methodology

 

During the first 18 months the project focused its efforts on gaining a thorough understanding of the scavenging poultry systems in the project locations, including the constraints affecting them. To do this it used a combination of methods.

 

Selection of villages and respondents

 

Trichy District, Tamil Nadu

 

The general project area was chosen partly on the basis that the state veterinary services were working closely with poultry-keepers here, and were interested in cooperating with the project team in the research; and partly because it was reasonably accessible from Namakkal, where the TANUVAS researchers are based. The TANUVAS team itself did not have a previous record of working on an ongoing basis in villages in Trichy, so the cooperation of the veterinary services was seen to be important in helping the team to establish a good rapport with the villagers.

 

Three categories of backyard poultry-keepers were identified in this district by the team in the early stages of the project. It was decided to work in two villages in which all three categories were present. This would enable us to be relatively confident that any differences found between the three groups could be attributed to the nature of their poultry-keeping systems, rather than other extraneous factors (e.g. distance from poultry market). Two villages were selected that satisfied this criterion, Peruganur and Ayyanar Kovil Salaikadu (AKS). The former is well connected to a small town and various services, whereas the latter is less well-connected.

 

The three categories of poultry-keepers identified were:

• Category 1 = small and marginal farmers whose home and poultry are adjacent to their agricultural land

• Category 2 = small and marginal farmers whose home and poultry are separate from their agricultural land, i.e. in a nucleated settlement

• Category 3 = landless people who live in a colony (hamlet), with poultry kept in and around the house.

Udaipur District, Rajasthan

 

The project team decided to work in three villages of Baghpura area in Jhadol block, since BAIF has a strong operational presence in this area, where it has been implementing a European Union-funded rural development project. The people living here are predominantly tribal, mainly belonging to the Bhil tribe. The project has been working through women's self-help groups (SHGs) in this project area.

 

Baseline survey

 

The project undertook a structured baseline survey of 60 poultry-keepers in Trichy district and 30 poultry-keepers in Udaipur district to obtain a general overview of practices and constraints (Conroy et al 2003). In Udaipur, 10 poultry-keepers were selected in each of the three project villages; while in Trichy 10 poultry-keepers from each of the three categories in each village were interviewed (except that in AKS only nine C3 poultry-keepers were interviewed as there were no more in this category). Most of the respondents were women, since they are usually responsible for all aspects of poultry-keeping.

 

The principal survey method was an interview schedule, which included a question about constraints and their relative importance. The enumerators were given the following instructions: and format for recording answers (Figure 1).

 

The enumerators were given the following instructions: Ask about poultry-keeping problems – do NOT read out the list below. First record answers with a tick. Then show interviewee cards symbolizing the problems they have mentioned, and ask them to rank the problems. The format used for recording answers is shown in Figure 1.

 

In addition, to collect information about hatchability and mortality the survey used a new technique, which we have called the participatory clutch history method (Conroy 2005). This information was obtained by getting the owner to recall what had happened to one or more specific clutches in her/his flock during the previous 6-9 months, and to record this information on a chart placed on the ground. Since many poultry-keepers are illiterate, the chart was based on symbols (e.g. use of stones to indicate numbers), rather than words and numerals. They would start (see Figure 2) by showing the number of eggs laid, then the numbers of eggs or birds at various subsequent stages, and ultimately the number reaching marketable age and retained in the flock. The 17 subsequent rows indicated possible explanations for removal from the flock, both deliberate (e.g. consumption, sale) and accidental (mortality due to disease, particular types of predator, etc.); and any removals were recorded in the appropriate row and column.
 

 

In each Udaipur village, a total of 15 clutch histories was taken from 10 respondents. In Trichy, 36 clutch histories were recorded in AKS and 42 in Peruganur.

 

Monitoring programme

 

After the baseline survey a one-year monitoring programme (from July 2001 to June 2002) was established in villages in the two locations to collect further information about poultry production and productivity. The programme, which covered 2056 birds in Udaipur and 1445 birds in Trichy, monitored weight gain and mortality and its causes. In addition, dead birds were examined for the presence of internal parasites: 94 birds were examined in slaughterhouses in Tamil Nadu, and 40 birds were sacrificed and examined in Udaipur at the state government's Regional Disease Diagnostic Centre.

 

Results

 

Baseline survey rankings

 

Trichy villages

 

Overall, predators and disease were perceived as being of roughly equal importance in Peruganur, as can be seen from Table 1. Predators were seen as the main problem by 12 poultry-keepers, and the second most important by nine; while disease is seen as most important by 11 and second most important by 10. Among category 1 and 2 poultry-keepers the two problems account for the vast majority of the top three rankings - 18 of the 20 top rankings and 16 of the 20 second rankings, respectively.

 

Among category 3 poultry-keepers feed-related problems acquire greater importance. This may be explained by the fact that C3 poultry-keepers are landless, and therefore they have to purchase grains; whereas C1 and C2 villagers have farms, and probably produce the majority of the grains given to their poultry.

 

In AKS, 28 of the 29 people interviewed ranked predators as the main constraint, while only one gave disease this ranking. Fifteen of them ranked disease as the second most important constraint, but 7 of the 9 C3 (landless) poultry-keepers ranked feed availability as the second most important.

 

Udaipur villages

 

In the Udaipur villages, as in Peruganur, predation and disease were perceived as being the most important problems. Roughly equal numbers of poultry-keepers rank them as the first and second most important problems: no other problems are ranked as first or second most important. Health service availability was ranked the third most important problem by 17 respondents, and feed availability by seven of them.

 

Baseline survey clutch history findings - overall mortality rates

The baseline survey's clutch history data (Conroy et al 2003) identified serious constraints on productivity in both locations. Respondents were generally aware of the causes of mortality. In the case of predation-induced mortality, they even knew the types of predators in most cases; although in a few cases they said they did not know, and in some cases they may have been guessing. Landless labourers may sometimes have not known whether a bird had been lost to a predator or had been stolen, given that they are away from their village during the day. Although reliability can be an issue when using recall methods (like the clutch history), the monitoring programme, in which causes of losses were recorded every two weeks, produced similar results. This suggests that recall reliability was not a problem.

Trichy villages

The baseline survey (Conroy et al 2003) identified serious constraints on productivity in both locations. In Trichy, losses were greater in the remoter village, AKS, where the overall mortality rate was more similar to that of the Udaipur villages than that of the second Trichy village, Peruganur (see Table 3). In both locations the survey data showed that predation was a more important cause of mortality than disease. The vast majority (95%) of predation deaths occurred during the first 8 weeks after hatching.

Udaipur Villages

The results are summarised in Table 4. The mean number of deaths per clutch was highest in Jagannathpura. Almost 42% of the birds that hatched died before reaching market age.

Mortality rates are similar in the three villages, but Richawar's is slightly higher than the other two. Overall, predation (21.9%) is a more important cause of mortality than disease (16.6%), and the predation rates in the three villages are quite similar.

 

Differences in Mortality Rates between Groups in the Trichy Villages

 

In Peruganur, disease is highest in category one birds and lowest in category 3, whereas predation mortality is similar (11 or 12%) for category 1 and 2 poultry-keepers, and highest (21.4%) for category 3 (see Table 5). In AKS, however, predation mortality rates are very similar for all three categories. The reasons for these differences are not readily apparent.

Baseline survey clutch history findings - types of predators

 

In Peruganur, all predation mortality was attributed to wild birds (see Table 6). The main type of predator for all three categories was large birds of prey (LBoP). while the second was small birds of prey (SBoP). Crows (Corvus spp.) were the third type of predator, but were involved much less frequently. There are two widespread crow species in India that are found near human habitation, namely the House crow (Corvus splendens) and the Large-billed crow (Corvus macrorhynchos).  In AKS, wild birds also caused more mortality than other types of predator, and again Large Birds of Prey is the main type.  However, in AKS crows are more important than Small Birds of Prey, whereas the opposite is the case in Peruganur. The principal mammalian predator is the wild cat, which is thought to be the Jungle Cat (Felis chaus). This has been described as "the common wild cat of India", and has been noted as preying on poultry (Prater 1980).

It was sometimes impossible to know precisely what species villagers were referring to when talking about birds of prey, so the researchers decided to use two general categories, Large Birds of Prey and Small Birds of Prey. The former could include kites, harriers, buzzards and eagles; and the latter could include Accipiters (which include sparrow hawks - the most common Accipiter in peninsular India is the Shikra (Accipiter badius), which is often seen near human habitation) and falcons. Information about birds of prey and their distribution in India has been obtained from Grimmett et al (1999).

 

Table 7 summarises the data on types of predators for the Udaipur villages. It appears that by far the most important predator is the crow, which kills more chicks than all the other predators combined. The mongoose (Herpestes spp.- probably Herpestes edwardsi) was also a significant predator, particularly in Richawar, and wild cats were the third most important.

 

Monitoring Programme

 

Mortality data from the monitoring programme in Trichy (Table 8), which was carried out in Peruganur, were generally consistent with those obtained through the baseline survey in Peruganur. The overall mortality rate was quite similar, and predation was more important than disease, but the gap (14.1%) between predation and disease mortality rates was much greater than that found by the baseline survey (7.1%).

A disaggregation of the 263 predation deaths (see Table 9) reveals that bird predators were far more important than mammalian predators, and that Small Birds of Prey (SBoP) were the main sub-group. There were also some striking differences, including the following:

• predation deaths were twice as many in the January-June period as they were in July-December

• the percentage of 'don't know predator' was far higher in the January-June period

• 88.2% of crow predation occurs in the July-December period

• more than 80% of mammalian predation occurs during the January-June period.

The higher number of deaths in Jan-June may be related to seasonal variations in the number of birds in the flocks.

 

Informal discussions with poultry-keepers

 

The research team in Udaipur convened a group discussion with poultry-keepers in Richawar village to discuss the predation problem and possible solutions. The discussion was based partly on the construction of a seasonal calendar by the poultry-keepers (see Table 10). It shows that: mammal predation is dominant in the rainy season; mammal and bird predation are equally important in the winter season; and bird predation is more important in the summer season.

The villagers said that predation is worst during the rainy season, so they deliberately reduce their flock size before the rainy season by consuming or disposing of their birds. The principal factor behind this is the amount of vegetative cover available to mammalian predators, which is greatest at this time: as cover decreases, so does predation by mammals. Conversely, bird predators are more of a threat when there is little or no cover available.

 

The figures given for the summer season are for someone who has no chicks (i.e. the majority of people): for someone with chicks, the figures would be 14 and 2. Of the 14 annas, typically 10 would be deaths caused by mongoose and the other four by wild cat. The Indian Fox (Vulpes bengalensis) is another significant predator. The figures for the winter season are for someone who is not growing wheat (only 20% do). Someone with wheat would experience slightly worse predation by mammals, due to the cover it provides.

 

The villagers confirmed that most predation occurs during the day, when the birds are outside. They said that the mongoose is the main predator during the rainy season, and tends to attack birds near the house: it takes birds that are up to 2-3 months old. The mongoose strikes mainly in the morning, the wild cat mainly in the evening, and the fox at any time of day. The wild cat takes birds that are up to one year old.

 

Crows only feed on chicks, up to about one month of age. Thus, they are the main bird predator in the winter season when the maximum numbers of chicks are produced. They are also rearing their own chicks at this time, and take the poultry chicks to feed to them. The villagers said that in the summer season the Black kite (Milvus migrans) is the main bird predator. It may attack 2-3 times in a day, whereas the eagle (species not known) only attacks once a day.

 

Discussion and Conclusions

 

Variations between villages

 

In both Trichy and Udaipur, predation was found to be a more important cause of mortality than disease, and the mean rates for the two districts were quite similar (23.3 percent and 21.9 percent respectively). There were, however, considerable differences between villages in predation induced mortality rates. The two lowest rates were 11.8 percent in Jagannathpura, Udaipur and 14.7 percent in Peruganur, Trichy; and the two highest rates were 24.3 percent in Richawar, Udaipur, and 31.8 percent in AKS, Trichy.

 

All predation is by birds in Peruganur; whereas in the other four villages, birds are the main predators, but mammals are also involved (and snakes in one case). There are also differences between villages regarding the importance of different types of birds of prey. Predation induced mortality rates must depend, at least in part, upon the presence and prevalence of predators, and this in turn will depend on whether there are suitable habitats (e.g. nesting sites for birds, sites for dens for wild cats, other feed sources) for them nearby. For example, AKS is situated adjacent to a hilly area, and before carrying out the baseline survey the project team anticipated that predation-induced mortality might be higher in this village than it had been in Peruganur, hypothesizing that numbers of mammalian predators and birds of prey would be higher in this area. Forests are another habitat that could harbour certain types of predator, such as jackals (Canis aureus) and birds of prey requiring tall trees in which to nest.

 

The Udaipur villages are also located in a hilly area (the Aravali hills), whereas Peruganur (which had a low predation mortality rate) is not located in close proximity to hills and is only six kms from a small town. The fact that there was no mammalian predation recorded in Peruganur (unlike AKS) may be a reflection of mammals not living sufficiently near to the village to be able to access it during daylight hours.

 

Another factor could be settlement patterns. For example, people and chickens living near the centre of a large nucleated village (i.e. where houses are concentrated in one place) might be less exposed to certain types of predator than those living on the periphery of a village, or people in isolated houses surrounded by fields. In Udaipur, the research was done with members of the Bhil tribe, and their normal settlement pattern is one in which a house "is erected on a small hillock in the midst of its patch of cultivated land" (Doshi and Vyas 1992).

 

Variations within Peruganur

 

Predation is highest in category 3, while disease mortality is lowest. One possible explanation for the high predation-induced mortality rate is that most of these poultry-keepers (both men and women) are landless labourers, who spend the day away from their homes. Thus, there may be less adults around to frighten predators away or to protect the chicks. A possible explanation for the low rate of disease-induced mortality is that predators may consume diseased birds without the owner (who is absent) knowing the birds had been diseased in the first place.

 

On the other hand, birds belonging to Category 1 poultry-keepers could be more exposed to predators, since they are surrounded by fields, and hence potentially more vulnerable to mongooses, wild cats, snakes etc. Category 2 poultry-keepers are similar to those in category 3 in that they live together in small colonies of houses, but perhaps different in that the women may be more likely to stay at home during the day.

 

Comparison with other studies

 

Our finding that predation causes more mortality than disease seems to be a unique one for India, and some people have expressed surprise at it. However, this is not the first study in India to note that predation is a serious constraint. A DANIDA-funded livestock development project in Koraput, Orissa, found that predation was 'an important problem', and noted that the main predators were crows, foxes (Vulpes spp.), Striped hyenas (Hyaena hyaena) and wild cats (Felix spp.) (Das et al 2003). The villages in this study were located inside forests or adjacent to forests, and wildlife legislation does not allow control of wild animals in the villages (ibid.). This study did not say how important predation was relative to other constraints.

 

It has been reported that in Madhya Pradesh predation is the second most important cause of mortality, after Newcastle disease (ND) (Mohapatra 2003). Another survey, conducted in five districts of the tribal belt in Western India (along the interstate boundaries of Rajasthan, Madhya Pradesh and Gujarat states) concluded that disease (especially ND) was the main cause of mortality, followed by predation (including theft) (Rangnekar and Rangnekar 1999). Finally, a fourth study, carried out in three districts of Madhya Pradesh, notes that predation was a cause of mortality in all three districts, but does not say how important it was compared with other causes of mortality (Kumtakar and Kumtakar 1999). However, none of these references provides quantitative data on the relative importance of different causes of mortality, and it is not readily apparent in most cases whether their assessments of the relative importance of different constraints are based on such data. In at least one of the studies such data were not collected (D V Rangnekar, personal communication).

 

It could be argued that the quantitative data on mortality presented in this paper are for a limited period of time, and that disease-induced mortality (particularly mortality due to Newcastle disease) may periodically be higher than what was observed in this study. However, in the three Udaipur project villages there have been no outbreaks of Newcastle disease during the four years that the project has been working there.

MD serology (HI test) was done on 151 samples from three villages, none of which showed the presence of antibodies against ND. This indicates that the birds had had no exposure to ND and had not been vaccinated against it (Bhardwaj and Bhatnagar 2004).

 

Predation control

 

Given the importance of predation-induced mortality, predation prevention measures deserve more attention from poultry research and development professionals. 'Improved housing' is the solution usually proposed by livestock scientists (Night shelters have been recommended as a means of reducing predation (Kumtakar and Kumtakar 1999), but clearly they would be ineffective if predation was only during the day).. However, since most predation takes place during daylight hours, when the birds are scavenging, for it to be effective scavenging would have to cease. This would require the owners to bring feed to the birds, thereby incurring labour and/or cash costs, which poor poultry-keepers might consider to be undesirable or not feasible. A bio-economic modelling exercise (intended to provide ex ante evaluation of interventions), which assumed 23-27% predation mortality before intervention and 0% after, found that daytime housing resulted in a large decrease in net returns (Udo et al 2002).

 

Fortunately, keeping birds in confinement is only one of several possible measures. Others suggested by poultry-keepers in Udaipur include providing cover (e.g. a bunch of thorny branches on the ground) for chicks against crows. Three measures suggested for dealing with the mongoose were:

• destroying the burrows of mongooses near to the home.

• burning chillies in the burrow - the smoke from them is apparently toxic; and

• before the birds are let out in the morning, throwing stones into the vegetation near the house, to scare away any mongooses that might be there.

Apart from daytime confinement, predation prevention measures tend to be species- or genus-specific, and they might also vary from one location to another. Thus, research is needed into problems and solutions in different locations.

 

References

 

Bhardwaj B and Bhatnagar C S 2004 Influence of Parasitic Infestation in Desibirds - Experiences in Udaipur. Paper presented at the 3^rd National Seminar on Rural Poultry for Adverse Environment, 25-26 February 2004, Tamil Nadu, India.

 

Conroy C 2005 Participatory Livestock Research: A Guide. London: ITDG Publishing.

 

Conroy C, Sparks N, Chandrasekaran D, Ghorpade A, Acamovic T, Pennycott T, Natarajan A, Anitha K, Pathan, R L and Shindey D N 2003 Key findings of the baseline survey on scavenging poultry in Trichy and Udaipur districts, India. Ayr, Scotland: Scottish Agricultural College.

 

Das K, Biswal A, Khuntia A, Parida J and Kar G C 2003 Status of village poultry production in Koraput district - ILDP Experiences'. Pp 14-26 in: State level workshop: poverty alleviation through poultry production in Orissa, 15-16 December 2003, Bhubaneswar, Orissa. Fisheries and animal resources development department, Government of Orissa, and Indo-Swiss Natural Resources Management Programme.

 

Devegowda G, Yathinder P V, Krishnamurthy T N and Kumar V (editors) 2002 Second National seminar on rural poultry for adverse environment, Bangalore 2002, 11^th-12^th December 2002. Bangalore: University of Agricultural Sciences.

 

Doshi S L and Vyas N N 1992 Tribal Rajasthan: Sunshine on the Aravali. Udaipur, Rajasthan: Himanshu Publications.

 

Grimmett R, Inskipp C and Inskipp T 1999 Pocket guide to the birds of the Indian Subcontinent. New Delhi, India: Oxford University Press.

 

Kumtakar V A and Kumtakar P 1999 Rural family poultry scenario in tribal areas of central Madhya Pradesh, India: a socio-economic analysis. Free Communication No. 9. First INFPD/FAO electronic conference on family poultry production. http://www.fao.org/ag/againfo/subjects/en/infpd/documents/econf_scope/add_paper9.html

 

Mohapatra S C 2003 Family Poultry Production in Tribal Belts of Madhya Pradesh and Maharashtra'. Pp 34-36 in: State level workshop: Poverty alleviation through poultry production in Orissa, 15-16 December 2003, Bhubaneswar, Orissa. fisheries and animal resources development department, Government of Orissa, and Indo-Swiss natural resources management programme.

 

Prater S H 1980 The book of Indian animals. Bombay Natural History Society and Manzar Khan Oxford University Press (Chennai).

 

Rangnekar S and Rangnekar D V 1999 Family poultry production in tribal belt of Western India. Free Communication No. 5. First INFPD/FAO electronic conference on family poultry production. http://www.fao.org/ag/againfo/subjects/en/infpd/documents/econf_scope/add_paper5.html

 

Udo H, Asgedom A and Viets T 2002 Modelling the impact of interventions in village poultry systems. In: Livestock, Community and Environment (Editors: Kyvsgaard N C and Monrad J) Proceedings of the 10^th International conference of the Associations of Institutions of Tropical Veterinary Medicine, AITVM, Copenhagen, pp. 237-246.