Birds nesting survival in disturbed and protected Neotropical savannas

Abstract

Fragmentation and other habitat disturbances are long known to negatively affect birds, in large part by decreasing nest success due to high nest predation rates. The factors, however, that cause this decrease in nest success are still poorly understood and may vary among regions or species. Here, we show that nest survival is also lower in a disturbed landscape versus a protected cerrado (savanna-like) Neotropical landscape. Also, we tested the importance of garbage in the nest, brood parasitism, microhabitat and bird family in nest survival, controlling for temporal effects. We monitored 144 birds' nests in a disturbed landscape and 150 nests in a natural reserve of cerrado vegetation in central Brazil, between September and December 2006. We used Program MARK to estimate nest survival probabilities and evaluate the effect of covariates in nest success in the disturbed area. Nest daily survival rate (DSR) was higher in the reserve (survival probability = 29.4%) than in the disturbed landscape (survival probability = 16.6%). Nest daily survival rate (DSR) was smaller in nests with garbage (survival probability = 9.3%) than in nests without garbage (survival probability = 19.5%) in the disturbed landscape. Effects of habitat disturbance on nest survival differed among bird families, with finches and tanagers being more affected mostly due to high nest predation rates. Conservation and management of birds in disturbed landscapes should include actions to decrease nest predation. In poor rural or suburban areas in developing countries, such as Brazil, actions like better garbage treatment may help conserve birds in disturbed landscapes.

Striped-tailed Yellow-finch nesting success in abandoned mining pits from central Brazilian cerrado

“…Our results are contrary to what has been found in other studies in the central cerrado region (Santos, 2008;Borges and Marini, 2010) and other seasonal environments in the temperate zone (e.g. Hochachka, 1990;Verhulst et al, 1995).…”

Predation on artificial nests in open habitats of central Brazil: effects of time and egg size

“…A previous study showed similar tendency for large eggs (quail eggs; França & Marini 2009a). Surprisingly, studies of natural nests in the same region found declining nest survival rates as the breeding season progressed (Borges & Marini 2010, Santos 2008. Artificial nests may be inaccurate for providing absolute nest predation rates on real nests, but may be sufficient for relative comparisons (Major & Kendal 1996, Buler & Hamilton 2000, Dion et al 2000.…”

“…In contrast, in some areas in the Cerrado, birds were the main predators (França et al 2009). Overall, predation in the Cerrado habitats tended to increase throughout the reproductive period (Borges & Marini 2010), decrease during nest development (França & Marini 2009b), and differ between habitats and microhabitats (França & Marini 2009a, Borges & Marini 2010. These studies show the importance of predation for population dynamics of neotropical birds and so experiments with artificial nests may be important to understand spatial and temporal patterns of reproductive success.…”

“…This is also true in open areas of the Cerrado (neotropical savanna), where predation results in 70% from 90% of nests lost (Lopes & Marini 2005, França & Marini 2009b, Borges & Marini 2010. In forests, the main nest predators appear to be mammals, snakes and lizards (Robinson & Robinson 2001, Stake et al 2004, Robinson et al 2005.…”

“…The tested hypotheses were: (1) predation on artificial nests with large eggs will be less than that on nests with small eggs (Roper 1992, Alvarez & Galleti 2007; (2) the risk of predation will increase over time during the reproductive season, as observed in some studies with natural nests in neotropical areas (Borges & Marini 2010, Duca & Marini 2005; (3) predation risk will be greater in shrubs habitat than in trees, because predation tends to be greater in savanna than in forest (see Duca (Dinsmore et al 2002), and to sort candidate models based on Akaike Information Criterion adjusted for small samples (AIC c ; Burnham & Anderson 1998). The best-fit model was the one with the lowest AIC c value and models with ∆AIC c ≤ 2 were considered to have substantial support to explain variation in the data (Burnham & Anderson 1998).…”