Modelando impactos da precipitação e maré em alagamentos de ninhos de bicudinho-do-brejo Formicivora acutirostris (Thamnophilidae)

• Autor(a) principal: Prado, Renan de Paula Ramos do
• Data de Publicação: 2022
• Tipo de documento: Trabalho de conclusão de curso
• Idioma: por
• Título da fonte: Repositório Institucional da UNESP
• Texto Completo: http://hdl.handle.net/11449/216647
• topic: Ecologia; Mudanças climáticas; Aves; Ninhos; Bicudinho-do-brejo

The Marsh Antwren (Formicifora acutirostris) is an endemic bird from southern Brazil, which inhabits coastal marshes exclusively. The species is classified as near threatened of extinction and has an estimated population of only 7511 mature individuals. Marsh Antwren build their nests in herbaceous vegetation in marshes regions that are periodically flooded by tide. The nests are a few centimeters from the ground and are susceptible to flooding caused by tide. Most marshes were formed 3,000 years ago, at the time the rate of sea level rise was 0.2 to 1.6 mm/year. Currently, the Intergovernmental Panel on Climate Change, through its sixth report, indicates that the rate of global mean sea level rise is about 3.7 mm/year. It is also increasing the global average temperature and extreme events such as precipitation and cyclones. Global warming is altering the distribution, abundance and other life-history features of species and increasing their risk of extinction. Thus, the objective of this study was to evaluate the effects of rains and tides on the number of nests lost by flooding, aiming the analysis of the Marsh Antwren viability. The effect of the accumulated rainfall, the maximum heights of the daily tides and the interaction of these two variables on the probability of occurrence of flooding of the Marsh Antwren nests was tested. We created a model using binomial regression with a generalized linear mixed model (GLMM), considering as fixed variables the accumulation of precipitation during five days, the maximum height of the daily tide and the interaction between them. We considered the years and the cycle as random variables. We observed that the effect are significant among all fixed variables, with significance levels p<0.05. However, the model presented deficiencies, not being able to make predictions of flooding for low and high values of effective probability of flooding.