Aspects of the biology of the chestnut-backed sparrow-lark (Eremopterix leucotis) in the Limpopo Province, South Africa

Abstract

Sparrow-larks form a relatively small genus in the family Alaudidae and comprise only seven species distributed widely throughout Africa and parts of the Eurasian landmass. Sparrow-larks are unique amongst larks in that they are sexually dichromatic and exhibit biparental care. The chestnut-backed sparrow-lark Eremopterix leucotis is endemic to Africa with five subspecies recognized based on differences in plumage colouration. The five subspecies are distributed throughout the arid to semi-arid savannas of Africa with two subspecies (E. l. hoeschi and E. l. smithi) occurring in southern Africa. Despite their widespread occurrence and its interest for research on the evolution of characteristics in the family (e.g. being sexually dichromatic and exhibiting biparental care), very little is known of the biology and ecology of the Eremopterix larks. The chestnut-backed sparrow-lark is no exception and most of what we know of the species is based on incidental observations from a few nests. In an attempt to improve our knowledge of this interesting group of species, it was decided to study various aspects of the breeding biology and ecology, moult, vocalizations and geographical variation in the chestnutbacked sparrow-lark. The breeding biology of the chestnut-backed sparrow-lark was studied at Al3 farm (De Loskop) near Mogwadi in the Limpopo Province of South Africa from January 2008 to December 2010. Data collected during the study included: breeding seasonality, egg and clutch characteristics, duration of the incubation and nestling periods, nest-site characteristics, the roles and relative contribution of the sexes in the breeding cycle, nestling development, diet and nestling provisioning rate, and breeding success. Chestnut-backed sparrow-larks bred mostly during the dry season, which is from April to September in the study area. Nevertheless, the results revealed that breeding is bimodal with a main peak in breeding activity in late summer and autumn (March to April) and a second smaller peak in spring (September to October). The species showed geographical variation in clutch size with a mean of 1.88 eggs recorded in the study area as opposed to 1.00 recorded in the northern parts of its range. Egg dimensions compared well with measurements obtained from the Nest Record Card Scheme of the Animal Demography Unit, University of Cape Town, South Africa. The mean incubation period of 10.33 days recorded in this study compares favourably with that of other Eremopterix species viii (8–10 days), a genus with some of the shortest incubation periods amongst larks. The mean nestling period of 9.2 days (range: 8–10) in the study area was significantly less than the 10–12 days reported for populations in the northern range of the species, but it compares well with those of other sparrow-larks. Nest site characteristics, which were quantified within a 1 m2 quadrant with the nest as the centre, including nest dimensions, were consistent with those reported in the literature. Chestnut-backed sparrow-larks in the study area preferred to nest in areas with a high percentage of bare ground (median = 67.5%) and very little vegetation cover (median = 25%). Most nests faced in a southerly direction compared to nests in the north of the species’ range, which face in a north-easterly or easterly direction. The species’ preference to face the nests away from the midday sun most probably serves a thermoregulatory function to avoid excessive heat during the warmest parts of the day. Most nests (78.2%) had an apron varying in size from small and insignificant to large and well-developed. The functional significance of the apron remains a matter of conjecture and there was no association between breeding success and presence or absence of the apron. In addition, one pair constructed one nest with and another without an apron, suggesting that individual preference or characteristic is not a determinant factor in the construction of an apron. Both sexes took part in nest construction, incubation and feeding and brooding of nestlings. However, the relative contributions were not entirely symmetrical as males incubated a greater proportion (50.1%) of the time compared to females (43.1%), and the mean and median of male incubation shift lengths were longer than that of females, albeit not statistically significant. However, females made statistically significantly (P < 0.05) more nest visits to deliver food compared to males (54.6% vs. 45.5%). The average breeding success, estimated using Mayfield’s method, was 16.1% but there were inter-annual differences with the overall breeding success in 2010 being only 8.1% compared to 20.6% of 2008. Known causes of failure included nest depredation, flooding, starvation, nest abandonment and hatching failure. Statistical analysis of morphometric data of live specimens and museum study skins suggest that, in addition to being sexually dichromatic, chestnut-backed sparrowlarks also exhibit mild sexual size dimorphism. However, there was considerable overlap in these measurements between the sexes and as a result the biological significance of this sexual size dimorphism may be negligible. Nevertheless, the ix results show chestnut-backed sparrow-lark males tend to have longer wings and tails compared to females. This may be adaptive with respect to the extended display flights that males perform during the breeding season. Interestingly, the mean mass of breeding females in the study area was significantly more compared to males, whereas the SAFRING database, representing data recorded throughout the year, showed no significant differences in the mean mass between the sexes. The greater mass of breeding females may relate to physiological changes associated with the acquisition of resources and the development of structures and tissues associated with egg-laying and egg-production. Larks rely heavily on vocalizations to attract mates and advertise territories. Not surprisingly, the study revealed a rich vocal repertoire for the chestnut-backed sparrow-lark. The analysis of the vocalizations shows that chestnut-backed sparrowlarks have a display song performed by males, a sub-song sung by both sexes and various different calls used in different contexts, e.g. flight and alarm calls. The study also presents the first analysis and description of the vocalizations of nestlings. An interesting feature of the vocalizations of the chestnut-backed sparrow-larks was that they performed hetero-specific vocal mimicry, which was incorporated in the subsong. Moult is a relatively unknown aspect in the annual cycle of the majority of larks. Chestnut-backed sparrow-larks undergo post-breeding moult, which is an adaptation to reduce the conflict between moult and breeding as both activities have high energy demands. The moult study also showed that they undergo a partial moult in mid-winter, involving the inner-most secondaries and some of the contour feathers. The results of this study shed valuable light on the natural history of this species and contributed significantly to ornithology and our growing understanding of the biology and ecology of the family. The results can also form a basis for future inter- and intraspecific comparative studies. The study illustrates the importance of undertaking long term studies of species to account for inter-annual differences in various ecological parameters.