Advanced search
Start date

Time series of algae community alterations in urban reservoirs: explanatory and predictive models in response to eutrophication and climatic changes

Grant number: 18/18896-4
Support type:Scholarships in Brazil - Post-Doctorate
Effective date (Start): February 01, 2019
Effective date (End): January 31, 2021
Field of knowledge:Biological Sciences - Ecology - Ecosystems Ecology
Principal Investigator:Denise de Campos Bicudo
Grantee:Jaques Everton Zanon
Home Institution: Instituto de Botânica. Secretaria do Meio Ambiente (São Paulo - Estado). São Paulo , SP, Brazil
Associated research grant:17/50341-0 - Challenges for biodiversity conservation facing climate changes, pollution, land use and occupation (PDIP), AP.PDIP


Understanding the past trajectory and the variability of aquatic ecosystems is fundamental to engender reference points as well as implement environmental recovery and management strategies, particularly in systems that have high ecological value and economic importance. Therefore, studies covering long time series are essential to avoid a reductionist view of nature that are usually unable to perceive the complexity of environmental problems in time and space. However, such studies are mainly carried out in tropical / subtropical regions of the globe, after the anthropogenic impact. Hence, a common problem for water quality management is the absence of long-term data that provides information on the pristine (or "baseline") ecosystem conditions, as well as the lack of a series of data collected in a continuous and systematized routine, which allows the elaboration of predictive models. In this sense, the area delimited is privileged because it integrates the effects of urbanization on forest and aquatic ecosystems. Using long time series from data obtained over 21 years (1997-2017), we will advance in understanding the effects of anthropogenic impacts and biodiversity changes, including climate change and artificial eutrophication. The use of spectral analysis allows the understanding of patterns at different time scales, which will drive the construction of predictive models. Using this analysis, time series will be 'dismembered' at different time scales, this enable us to find out which major temporal cycles are related to which exogenous signals. In addition, the use of multivariate auto regression models will allow us to explore the interaction structure of system´s species / groups, moving forward in the environmental diagnosis. This model allows us to quantify the degree of competition or facilitation within and between groups as well as some community processes (e.g. resilience). In addition, we will employ an ecological regime change detection methodology. We will use this approach combined with with pristine conditions (or "baseline") as a reference, this will allow us to assess the degree of system´s disturbance due to climatic oscillations and eutrophication. Due to the high complexity of the data, we study community ecology using distance-based ordination methods. Our approach is innovative because it provides information poorly explored in community ecology (e.g. spectral analyzes, structure of interactions, change of regimes) allowing us to manipulate information that is scarce in community ecology. This will be possible due to the particularity of the data to be used, for addition to a large time resolution (20 years), it is a regular time series (equally spaced: monthly).